Wanting to front-run the crowds, I made plans with a friend and fellow crypto enthusiast to head down to the exhibition on opening day.

Notes from the Ether is a digital art exhibition featuring works engaging with the emerging technologies of non-fungible tokens (NFTs) and generative AI. The exhibition boasts a solid line-up of 20 artists, including those who have essentially become household names within the NFT space, such as Tyler Hobbs and Larva Labs. Curated by ArtScience Museum curator Deborah Lim and guest curator Clara Peh, Notes from the Ether will run from 19 August 2023 to 24 September 2023.

Those who may be less familiar with NFTs and generative AI should not feel intimidated from visiting this exhibition, as the explanatory statements accompanying the displays should provide sufficient context for these technologies and how they are employed in the displayed works. That said, it will be prudent to be patient—as these works do push the boundaries of what we think art is and could be, it will take some time to appreciate them.

The exhibition is organised into different sections focused on a specific theme. Taken as a whole, the different themes reflect the range of artistic practices that blockchains and generative AI have made possible. The first section on "Coded Possibilities" celebrates artworks created with algorithms, such as Larva Labs' Autoglyphs—the prime example of a generative artwork whose code script was placed on the blockchain, thereby imbuing the resultant outputs with a certain immutability and longevity.

Another piece that is likely to catch your attention is Tyler Hobbs' and Dandelion Wist's QQL project, which allows collectors to play a role in the creation of generative artworks by offering them the agency to decide certain parameters in the algorithm before deciding which outputs to mint as NFTs. That said, interacting with the QQL algorithm is as much an exercise in control as in surrender. After all, a fundamental element in generative art is randomness, and the parameters you pre-determine are meant to scope but not eliminate this element of chance. Leave the parameters the same, and be surprised by the diversity of the outputs that can still emerge!

Capping off the first section, Emily Xie's Memories of Qilin series features organic forms and textures that hint at elements drawn from East Asian folklore—of dragons, phoenixes, flowers, and mountains. More fundamentally, it showcases the visual subtleties that code can evoke, demonstrating that words and numbers have the ability to call forth life-like compositions and patterns that defy their binary roots.

https://opensea.io/assets/ethereum/0xa7d8d9ef8D8Ce8992Df33D8b8CF4Aebabd5bD270/282000798

The second section, "Co-creation", introduces the role of AI in the process of artmaking, and showcases Botto as an example of a form of human-machine collaboration that endeavours to give as much agency as possible to the latter. Conceptualised as a "decentralized autonomous artist" by Mario Klingemann and the software collective ElevenYellow, Botto uses a custom text generator and text-to-image AI models to create thousands of new artworks each week. A “taste” model then selects 350 images to present to the community for voting, with the results determining the canonical piece to be minted as an NFT and auctioned off for the week. Beyond the co-creation angle, I would add that Botto's works are also beautiful because they offer you a chance to see a machine dream on its own terms.

https://opensea.io/assets/ethereum/0xb932a70A57673d89f4acfFBE830E8ed7f75Fb9e0/33879

At the risk of spoiling the rest of the exhibition, I shall not elaborate on the rest of the sections in detail. Suffice to say, the remaining sections explore other key aspects of blockchain or generative AI technologies—such as interrogating the concept of ownership, contemplating the interplay between the digital and physical, as well as examining the formation of new online communities and networks.

These artworks may not be conventionally "aesthetic", but they certainly possess a conceptual depth whose artistic merits can only be appreciated with time. As blockchain and generative AI technologies become more pervasive, there will be a growing need to make sense of these technologies and carve out space for a healthy, human experience in the increasingly digital landscapes we will inhabit. This is a key role that art has to play, to smoothen the inflection point ahead of us, from which the boundaries between man and machine will become further and irrevocably intertwined.

So dive into the ether through this exhibition, and do not be surprised if you emerge from the other side with a newfound appreciation for the creative possibilities of technology, and how it can ultimately enhance the human condition.

This essay was first published on Joyn on 20 August 2023.

The next morning, I was greeted with this email from SuperRare:

With that, I had to try to rationalise what went on in my head last night. After all, who would in the right mind would trade a significant portion of one’s liquidity with a jpeg image that anyone could right click and save as?

https://opensea.io/assets/ethereum/0x1C7576619032eAf8B8A938C352E535bbA92a366c/11

Botto’s caption: We're clubbing with multi-American maniacs, viewed and depicted in their decadent ensambles. Socializing with sonic vitality, movement, laughter and frivolity. A melange of stories, cultures and hues, living loud and proud without any dues. No need for permissions or reservations, escaping all the limitations and restrictions, crazy and chaotic, living life in fast-forward, and dancing like there's no tomorrow's border.

A timely work

Good art is often regarded as timeless. However, as much as we want art to transcend the passage of time, we also want it to engage with the present. After all, timeliness can also be the wellspring from which timelessness ensues. In this regard, I view Clubbing with Multi-American Maniacs as a timely work—one which I hope will serve as a meaningful reflection of this particular moment in time, not only for the broader digital art community, but also personally as an individual.

While many artists have been experimenting with generative AI technologies to create art, Botto stands out in that it does so without direct human input as a “decentralised autonomous artist”. Botto’s AI-powered art engine comprises a custom text generator, which creates text prompts that are sent to its text-to-image models to generate up to 8,000-10,000 images per week. A “taste” model then selects 350 images to present to the community (i.e. the Botto DAO) for voting, with the results determining which is Botto’s canonical piece to be minted as an NFT and auctioned on SuperRare for the week.

With no human allowed to interfere with the image generation process (besides providing feedback through voting), the intent is for Botto to pioneer an entirely novel approach to artmaking, which emphasises providing maximum agency for a machine to dream on its own terms. I believe this is an innovative and timely experiment in digital art, especially since we are on the cusp of witnessing fundamental shifts in our economy and culture due to generative AI technologies. Hence, I wanted to own a work from Botto now, as I wanted to be part of this history—to mark out my awareness of this potential inflection point, from which the boundaries between man and machine will likely become further and irrevocably intertwined.

Reflecting the current state of the world

The key thing that stood out to me with Clubbing with Multi-American Maniacs was that it is the first minted work by Botto that referenced a real-life political entity in its title, in this case, the United States (US). With this in mind, it is therefore only natural to perceive this work in the context of the current socio-political landscape of the US and the world.

Seen in this light, the cheery rambunctiousness in the work seems to reflect a brazen disregard for any semblance of propriety, as the figures descend into a wanton and unhinged brawl. Botto’s caption seems to convey as much: “No need for permissions or reservations, escaping all the limitations and restrictions, crazy and chaotic, living life in fast-forward, and dancing like there's no [tomorrow].” In fact, this was precisely the impression that the US left on the rest of the world during the Trump administration, when it had little qualms upending international norms while having no compunction letting its internal divisions seethe and boil over for the world to see.

That the scene in Clubbing with Multi-American Maniacs is filled with only old white men also hints at the dynamics of the US’ evolving politics, in particular, the waning influence of the old guard. Indeed, the White Anglo-Saxon Protestant (WASP) establishment is being increasingly challenged from within and without. More than just a deepening polarisation between left and right, the US seems to be splintering into what the American journalist George Packer has categorised as four rival narratives—”Free America”, “Smart America”, “Real America” and “Just America”. The tensions between these four camps can be fierce, as Packer describes:

They all anoint winners and losers. In Free America, the winners are the makers, and the losers are the takers who want to drag the rest down in perpetual dependency on a smothering government. In Smart America, the winners are the credentialed meritocrats, and the losers are the poorly educated who want to resist inevitable progress. In Real America, the winners are the hardworking folk of the white Christian heartland, and the losers are treacherous elites and contaminating others who want to destroy the country. In Just America, the winners are the marginalized groups, and the losers are the dominant groups that want to go on dominating.

“Clubbing” in the context of Botto’s work thus cannot refer to a fun night out, but has to contain a more foreboding undertone. Perhaps, the work reflects a final spurt of decadence after a gilded age—the contemporary equivalent of the Roaring Twenties—before the US confronts the sharp peak of instability that the American academic Peter Turchin has predicted may threaten to pull the country apart in the 2020s. Surely, with the risks of AI further deepening and entrenching inequalities with American society (and in most other parts of the world), the prospect of polarised factions clubbing each other to a maniacal end cannot be discounted.

Coincidence or not, I think it is noteworthy that Botto has managed to tap on the vast corpus of human-created data to generate a work that can reflect the zeitgeist of the times. In addition, considering that the votes of the community across several rounds would have played a part in training Botto’s taste model and scoping the latent space that Botto can operate in (through the theme of each Period), it should not be a stretch to say that Clubbing with Multi-American Maniacs represents the gaze of the machine as turned back onto us. One then has to ask: do we like what we see?

My personal connection with America

I collected Clubbing with Multi-American Maniacs not only because it could serve as a fodder for some academic commentary on contemporary American society. More than that, I feel that I can relate to the work on a personal level—conceptually and experientially.

Even though I am not American, I believe the US has had a strong influence in my life. As a millennial, I grew up in Singapore in the post-Cold War period, when our economy grew strongly amidst the globalisation championed by the US. As kids, we also consumed copious amounts of American media—news, music, movies, TV shows, games. While we may not be considered Americanised, we were familiar with American culture, or at least popular depictions of it.

More personally, I went to college in the US and spent the formative years of my early adulthood there. Although I will not consider the US home, it has left a deep imprint on me, which I am still trying to make sense of. I guess dealing with the alienation of being a foreigner while also figuring out my own identity in the transition from adolescence provided a fertile ground for me to imbibe certain supposedly American characteristics—an irrational optimism towards the future and a growing appreciation of diversity. But not all was well, and there were many instances when I felt lonely and unable to connect with my peers, and also simply disillusioned with the superficiality of college life.

Looking at Clubbing with Multi-American Maniacs, I am reminded of all these memories and emotions of my time in the US. On balance, this was definitely a period of significant personal growth for me, and I must credit my experience as a liberal arts student in the US, which encouraged me to take on a much more expansive view of the world. Although I certainly would not want to go clubbing like a maniac in the US, I do hope that by collecting this piece from Botto, I can—taking reference from Botto’s own words—relive the “sonic vitality, movement, laughter and frivolity” of my college years, and the impulse then to dive into the “melange of stories, cultures and hues” in American society and beyond.

Moreover, there is something symbolic about a non-American holding on to this seemingly US-focused artwork. Perhaps, there may be a little bit of the American spirit in all of us. Ultimately, we are all connected, and what happens within the US will have profound reverberations on the rest of the world. Indeed, I will constantly think about the “multi-American maniacs” depicted within, and what they portend for both the world and my own life.

In the meantime, to paraphrase from Botto again, I guess it won’t hurt to dance like there is no tomorrow, without permissions or reservations.

Generative art and post-photography are two exciting fields of art that have come to the fore amidst the groundswell of creativity in recent years brought about by NFTs and generative AI technologies. In both fields, man and machine are intimately entwined. Artistic expression is mediated by algorithms, which while created and honed by humans, have now the space and agency to transcend our creative vision and challenge our pre-conceptions.

William Mapan’s Strands of Solitude is a generative art series of 400 pieces, curated and launched on the Tonic platform. The series is framed as “a meditation on connection and isolation in the digital age”, borne out by its intricate compositions featuring highly-distorted elements that are nevertheless underpinned by a strong visual coherence. Though each piece is algorithmically generated, they all retain a palpable sense of humanness, reflecting a seemingly hand-made quality through the use of organic forms and rich texture.

Roope Rainisto’s REWORLD is a series of 500 works created using generative AI methods and launched as part of a larger post-photographic group collection titled “Post Photographic Perspectives” on the Fellowship platform. The series explores society’s complexities through the lens of the machine, depicting familiar subjects and settings in a distinctly non-human manner. Each piece appears to unravel our perceptions of reality, but also contains a glimpse of new, alternative possibilities.

Despite exhibiting starkly different visual profiles, Strands of Solitude and REWORLD converge in the conceptual terrain that they traverse. At the heart of both series lies the tension between boundaries and possibilities, which governs how each piece came to be and what they may become. Through the pairings below, this tension is played out across both series, and projected on broader questions about the individual’s place in society and the relevance of the human imagination in a world where machines too can dream.

Strands of Solitude #330 | Feeding Hand

The dense composition in #330, seemingly set in motion by the use of flow fields, reflects the circulation of entities and energy that drive human society.

As our urban landscapes trend towards the monumental, not unlike the scene depicted in "Feeding Hand", the amount of energy that must be consumed to power this expansion will surely be staggering.

Can we then afford to bite the hand that feeds us?

https://opensea.io/assets/ethereum/0x720786231DDF158EbD23bD590F73b29BFF78D783/330

https://opensea.io/assets/ethereum/0x66bcf059B2D8dB6cD10615Cdd781828db58268cF/165

Strands of Solitude #227 | Handmade Art

The human touch has always been an essential component of art. But as #227 hints at, what if machines can reproduce this hand-made quality, and do it better?

Is the scene in “Handmade Art” a foreboding of what would become of the human hand—an aimless side-show, reduced to making detritus?

https://opensea.io/assets/ethereum/0x720786231DDF158EbD23bD590F73b29BFF78D783/227

https://opensea.io/assets/ethereum/0x66bcf059B2D8dB6cD10615Cdd781828db58268cF/201

Strands of Solitude #385 | Symbolism Failure

As we train our algorithms on the world, they will have to perceive and interpret our landscapes, our constructions, and our symbols. #385 and “Symbolism Failure” suggest that they may do so in discordant, dissonant ways.

But how will we know what our algorithms really see?

https://opensea.io/assets/ethereum/0x720786231DDF158EbD23bD590F73b29BFF78D783/385

https://opensea.io/assets/ethereum/0x66bcf059B2D8dB6cD10615Cdd781828db58268cF/414

Age of Ads | Strands of Solitude #286

In our rapidly digitalising society today, it is almost impossible to escape the reach of advertising, exemplified by the cluttered billboards in “Age of Ads”.

But the hunt for profits can also be a tremendous source of creativity, holding the potential to dazzle and add colour to our lives, à la #286.

Where will this competition for attention take us, as we move into a world turbo-charged by AI?

https://opensea.io/assets/ethereum/0x66bcf059B2D8dB6cD10615Cdd781828db58268cF/5

https://opensea.io/assets/ethereum/0x720786231DDF158EbD23bD590F73b29BFF78D783/286

Strands of Solitude #52 | Age of Enlightenment

Nirvana is often portrayed as an elusive thing—an ungraspable state, a fleeting moment, like the burst of reddish hues in #52.

In “Age of Enlightenment”, it seems to come all of a sudden, as a soul-ripping rupture in one’s identity.

Can enlightenment itself be automated?

https://opensea.io/assets/ethereum/0x720786231DDF158EbD23bD590F73b29BFF78D783/52

https://opensea.io/assets/ethereum/0x66bcf059B2D8dB6cD10615Cdd781828db58268cF/6

Dual Mirror No. 1 | Strands of Solitude #162

The human gaze is limited within the space we have carved out within the natural world, as “Dual Mirror No. 1” suggests.

But #162 flips this state of affairs around, with its Echo Chamber characteristic conjuring an artificial boundary that seemingly constrains and subjugates a swath of nature.

What will our relationship be nature be like when everything becomes algorithmically-driven?

https://opensea.io/assets/ethereum/0x66bcf059B2D8dB6cD10615Cdd781828db58268cF/144

https://opensea.io/assets/ethereum/0x720786231DDF158EbD23bD590F73b29BFF78D783/162

Strands of Solitude #305 | Crisis Climate

The heavy black hues in #305 evoke the stain and weight of pollution in our natural world. But our environment can also fight back and stop our machines dead in their tracks, as “Crisis Climate” demonstrates.

Can man, machine and nature work better together to preserve this world we all call home?

https://opensea.io/assets/ethereum/0x720786231DDF158EbD23bD590F73b29BFF78D783/305

https://opensea.io/assets/ethereum/0x66bcf059B2D8dB6cD10615Cdd781828db58268cF/102

Strands of Solitude #192 | The Magician

The Conflict characteristic in the Strands of Solitude collection, which allows a second colour palette to appear within the piece, creates space for moments of magic to emerge from the algorithm.

The burst of grey tones at the heart of #192 is an example of this characteristic in play, with a sense of motion almost in sync with the bubbles conjured in “The Magician”.

What other magic remains hidden amidst the lines of code that underlie our algorithms?

https://opensea.io/assets/ethereum/0x720786231DDF158EbD23bD590F73b29BFF78D783/192

https://opensea.io/assets/ethereum/0x66bcf059B2D8dB6cD10615Cdd781828db58268cF/436

Where Are You | Strands of Solitude #134

“Where Are You” depicts seemingly identical individuals on the lookout for something, somewhere. They could very well be searching for a flash of inspiration out of all that monotony, like the dash of green hues in #134.

When everything we have is within the realm of algorithmic possibility, is there still room for us to seek the exceptional?

https://opensea.io/assets/ethereum/0x66bcf059B2D8dB6cD10615Cdd781828db58268cF/491

https://opensea.io/assets/ethereum/0x720786231DDF158EbD23bD590F73b29BFF78D783/134

All The Same | Strands of Solitude #339

In “All The Same”, the long shadows of the crowd create the impression that everyone is moving in the same direction, in unison towards the same end.

While this connotes the overwhelming influence of a herd mentality, #339 offers the perspective that cohesion and coherence can also have a transcendent beauty—where each individual part melds into a collective whole, aligned in flow and momentum.

Is there a singularity that our algorithms are leading us towards?

https://opensea.io/assets/ethereum/0x66bcf059B2D8dB6cD10615Cdd781828db58268cF/11

https://opensea.io/assets/ethereum/0x720786231DDF158EbD23bD590F73b29BFF78D783/339

Strands of Solitude #172 | Male Dominance

The flesh-coloured elements in #172, seemingly disjointed and tossed around, share some resemblance with the transmogrified bodies in “Male Dominance”. They certainly serve as a harbinger of the growing impact of generative AI in the intimate and private spaces where we satiate our erotic desires.

Will we still yearn for the human flesh in reality when we can generate its representations infinitely?

https://opensea.io/assets/ethereum/0x720786231DDF158EbD23bD590F73b29BFF78D783/172

https://opensea.io/assets/ethereum/0x66bcf059B2D8dB6cD10615Cdd781828db58268cF/260

Strands of Solitude #252 | Magic Hour

The gentle, relaxed tones in #252 evoke a similar vibe as the setting in “Magic Hour”—a state of tranquility far removed from the vicissitudes of the world.

When our algorithms are in a constant process of evolving, what sort of equilibrium can we expect to achieve?

https://opensea.io/assets/ethereum/0x720786231DDF158EbD23bD590F73b29BFF78D783/252

https://opensea.io/assets/ethereum/0x66bcf059B2D8dB6cD10615Cdd781828db58268cF/259

Tomato Colored Tracksuit | Strands of Solitude #258

The subject in “Tomato Colored Tracksuit” defies our social conventions, wearing an eye-catching orange attire and carrying a giant tomato in the middle of the countryside.

What invisible forces would have laid unseen here, repressing this vivid expression of the self?

Will AI set our individuality free, or as #258 seems to reflect, will it only strengthen the cages that society pushes down on us?

https://opensea.io/assets/ethereum/0x66bcf059B2D8dB6cD10615Cdd781828db58268cF/453

https://opensea.io/assets/ethereum/0x720786231DDF158EbD23bD590F73b29BFF78D783/258

Strands of Solitude #288 | Blue Lagoon

The crisp, distinctive swirls in #288, marked out by their blue and yellow hues, evoke the uplifting, heady feeling of heading on a vacation—perhaps to a beach on a warm, summer day like the one shown in “Blue Lagoon”.

Will we still feel this lightness of being when we know that our leisure activities are increasingly governed not by our own decisions, but by the unknown ways of our algorithms?

https://opensea.io/assets/ethereum/0x720786231DDF158EbD23bD590F73b29BFF78D783/288

https://opensea.io/assets/ethereum/0x66bcf059B2D8dB6cD10615Cdd781828db58268cF/44

Strands of Solitude #31 | Treading Water

The resemblance between #31 and “Treading Water” is uncanny. Both depict a human figure within a bounded space, amidst what appears to be a body of water. It is unclear whether they have done so of their volition or were kept there against their will. Regardless, this solitary state seems to provide an appropriate setting to contemplate one’s place in society.

But when we have to keep struggling to stay above the relentless cacophony of noise and data in the world, how can we preserve the space for solitude and introspection?

https://opensea.io/assets/ethereum/0x720786231DDF158EbD23bD590F73b29BFF78D783/31

https://opensea.io/assets/ethereum/0x66bcf059B2D8dB6cD10615Cdd781828db58268cF/460

Crescent Moons | Strands of Solitude #60

The illuminated wire structures in “Crescent Moons” sparkle against the night sky—evocative of the light rays propagating through the network of fibre optic cables that bind our society today.

But this connectedness also has the potential to tear us apart, scattering us like the bits and particles in frayed cables that #60 seems to conjure.

Connected or alone—in a world where our actions are determined by technologies beyond our control or even comprehension, what remains of human agency? As we continue charting a path within this momentary flicker of existence that is life, will our instincts, honed over millennia of human evolution, be eventually displaced by our algorithmic creations?

https://opensea.io/assets/ethereum/0x66bcf059B2D8dB6cD10615Cdd781828db58268cF/101

https://opensea.io/assets/ethereum/0x720786231DDF158EbD23bD590F73b29BFF78D783/60

As someone without coding experience, much less an appreciation for the technicalities and intricacies of software programming, the applications of code in any form have never seized my curiosity or imagination. I know that code forms the substrate of our increasingly digitalised lives, but I was content to simply remain a passive user.

However, around six months ago while nursing a COVID-19 infection, I stumbled onto the field of generative art, which can be loosely understood as art created from code. At that time, my fiancée and I were in the process of purchasing our new home together, and had just begun consulting our interior designer on the renovation works. With the copious amount of personal time while on quarantine, I started thinking about featuring elements from my growing interest in the crypto space within our new home. Somehow or another, I soon found myself spending more and more time looking at non-fungible tokens (NFTs) of long-form generative art.

Generative art and NFTs

Generative art is not a new phenomenon. It can trace its roots to artistic movements in the early 20th century, such as Futurism and Constructivism that introduced technology and industrial modes of production as key themes, as well as Dada and Surrealism that incorporated elements of spontaneity and randomness to challenge established norms. The term “generative art” itself was coined in 1965 by Max Bense, a German philosopher, to describe works that can be made through a set of rules, while retaining some degree of chance in the creation process.

As the influence of digital technology on society became more pervasive throughout the later half of the 20th century, more artists inevitably began to explore the intersection of art and computer science. This laid a firm foundation for digital art, including generative art which is often created via digital processes. (For a more thorough background on generative art, check out this excellent overview of the history and significant developments in the generative art space by Jason Bailey, who is behind the art and tech blog, Artnome.)

https://www.artnome.com/news/2018/8/8/why-love-generative-art

Nevertheless, digital art is often perceived as less relevant or worthy compared to physical art, perhaps due to the fact that it is less tangible (you cannot hold it in your hands) and is inherently replicable (you can easily “right click and save” a digital file indefinitely). The emergence of NFTs on blockchains in recent years has started to change this state of affairs, however.

To understand the impact of NFTs on digital art, we first need to be clear on what they are. In essence, NFTs are unique digital certificates that live on blockchains, with meta-data that links each token to a particular asset, such as an artwork, a music clip, or anything for that matter. So NFTs are not the artworks themselves. Instead, they are more akin to certificates that represent ownership over the artworks that they are linked to.

The game-changer with regard to NFTs is that the provenance of each token is immutable and can be easily verified by anyone, thanks to the design of blockchains. These qualities enable digital art, for the first time, to be considered scarce. While the underlying digital art piece can be reproduced infinitely, the NFT serves as the only authentic certificate linked to that art piece and it cannot be forged or tampered with. As a result, NFTs have dramatically expanded the market for digital art, allowing digital artists to better monetise their works while spawning a whole new generation of digital art collectors.

Long-form generative art

Specific to generative art, blockchain technology and NFTs have led to a new and dynamic genre of generative art, what is often referred to as long-form generative art.

As the generative artist Tyler Hobbs explained in this essay, the creation of generative art has historically involved a process of curation, in which only a single output from a particular generative algorithm, or a small set at most, is selected and showcased. However, blockchain-based platforms for generative art, in particular Art Blocks (on the Ethereum blockchain) and fxhash (on the Tezos blockchain), have popularised a different approach, where the artist places the underlying algorithm on the blockchain and pre-determines the number of iterations that can be “minted” as NFTs from the algorithm. The term “long-form” arises from the fact that collection sizes on these platforms are typically in the range of a couple of hundreds to 1,000, far greater than the norm for generative art projects before.

Another distinctive feature of long-form generative art is that the outputs of each iteration go directly to the collector, and the artist cannot intervene to decide whether a particular iteration should or should not be part of the collection. Once an iteration is created from the algorithm, it is linked to an NFT that will permanently and immutably live on the blockchain, and that NFT is transferred to the collector’s wallet.

The appeal of long-form generative art

Over the past couple of months, I have collected several works in long-form generative art projects, and I intend to collect more of them if my real-life wallet allows. Besides displaying a couple of these works as prints in my new home, I also regard some of them as assets that could serve as a good store-of-value, or even appreciate in value in the future, at least when denominated in their native cryptocurrency (i.e. ETH or XTZ). But perhaps more important than their monetary value, my intent is for the works I collect to have a certain longevity, and to continue to fascinate me (and hopefully others) over time.

To me, the appeal of long-form generative art can be outlined via three aspects: content, context, and community.

Content

Fundamentally, I regard art as a subjective experience, which is first and foremost informed by the content of the artwork—in effect, how it looks, sounds, or feels like, depending on the type of art. I do not think I can appreciate a piece of art if I do not intuitively like how it appears to my senses or manifests in my mind.

On this front, there are sufficient examples of long-form generative art projects that have captured my visual interest to make me fascinated about the genre as a whole. In particular, I find it incredible to witness how lines of code can generate compelling forms and patterns, reflected in compositions that span the gamut from being simply beautiful to those that are breathtakingly intricate.

One of my favourite long-form generative art projects is Meridian by Matt DesLauriers, which was launched as a 1,000-edition collection on Art Blocks in September 2021. As described in the artist’s notes on the project, Meridian is essentially “a 15kb JavaScript program that constructs stratified landforms using hundreds of thousands of small strokes of colour”, and the hash of each token in the project “describes a coordinate within a multidimensional generative space, locating a unique composition that lies along one of many possible longitudes.” Beyond any verbal description, however, I think the works in this collection have to be appreciated visually, both at a distance and when zoomed into the details.

Go to the project page on Art Blocks and scroll through the different iterations—notice the ways in which the different layers flow. They are simply sublime!

Context

Of course, having appealing content per se does not distinguish long-form generative art from other forms of generative art, or digital art more generally. Any art work can look nice. But what makes long-form generative art distinctive is its context, in terms of how it is created and therefore how it can be perceived.

As earlier mentioned, generative art refers to art that is created from a set of rules, with some element of randomness thrown in. In the case of Art Blocks, the algorithm for each project is stored directly on the Ethereum blockchain. When someone creates an output by minting an NFT, a random string of hexadecimal characters is generated and placed in the NFT, which then serves as the input “seed” for the algorithm and determines how that particular output looks and operates.

Understanding this context allows one to appreciate the output not only based on its physical manifestation, but also consider how it is the product of what is effectively controlled randomness. I find it fascinating that the creation process of generative art is embedded with a certain paradox—the artist uses a highly-deterministic medium, i.e. code, but yet incorporates the role of chance in determining the final outputs. In my opinion, this paradoxical creation process adds another layer of beauty to the works, reflecting how they are born from a delicate dance between rules and randomness. This is perhaps also testament to the symbiotic relationship between the artist and the collector, which is all the more pertinent in light of the role of blockchains and NFTs in facilitating closer relationships between both parties.

At the collection level, the important contextual feature of long-form generation art is the lack of curation on the artist’s part. As highlighted earlier, there is no opportunity for the artist to selectively feature or hide the outputs from the project’s algorithm. Each output after the project is launched becomes permanently and immutably part of the collection on the blockchain, limited only by the pre-determined collection size.

In this sense, each output in a long-form generative art project should not only be perceived as a standalone art piece, but also appreciated in relation to the entire collection. With long-form generative art, the range of outputs from the algorithm will be highly visible to collectors, which imposes additional demands on the project. Firstly, the algorithm must be able to produce outputs in a consistent and coherent manner at scale; a good long-form generative art project cannot have an algorithm that is tuned towards producing high-quality outputs for only a small subset of works, while generating garbage outputs for the rest. Secondly, there should still be enough variety across the outputs to justify the collection size, otherwise the collection as a whole may appear repetitive. As Tyler Hobbs has written:

“With a top-class algorithm, every output will have something new about it, a little surprise that teaches you more about what is possible.”

To me, this interplay of coherence and variety at the collection level provides a rich context to appreciate each individual output from a long-form generative art project. For example, I can look at each output based on specific parameters as programmed by the artist into the algorithm (e.g. the colour scheme, or use of specific patterns or elements), and think about how they manifest in different ways across the collection.

More broadly, I can also consider how the central theme in the project is expressed and pushed in different boundaries by that delicate dance between rules and randomness inherent in the genre itself. In fact, I would have to reckon that Charles Darwin’s poignant conclusion to his treatise on evolution would apply to the very best of long-form generative art as well:

“[F]rom so simple a beginning, endless forms most beautiful and most wonderful have been, and are being, evolved.”

Community

Finally, the relatively larger collection size of long-form generative art can also make the experience of collecting generative art much more social, as compared to projects with only a small selection of outputs that is showcased. Given that more collectors can have a stake in a long-form project, there will naturally be a larger collector base to tap on for community-building.

For long-form generative art in particular, the coherence of each project can serve as a common frame of reference to anchor a community of collectors, while its inherent variety can serve as fodder for continued engagement within the community. In fact, collectors or fans of a certain project often share their favourite outputs on social media, highlighting what makes them special in their eyes, and finding unique pairings or combinations with other outputs. Collectively, such sharing can sustain discourse within the community, and enrich everyone’s appreciation of the project as a whole.

https://twitter.com/punk6529/status/1439332605263032327

https://twitter.com/plutoniumfitzg/status/1551911764102909952

A number of platforms have also been developed which enable creators and fans to create their own NFT galleries. Of note, Deca allows users to curate galleries featuring NFTs that they do not own, while also giving them the tools to add custom backgrounds and text to present their galleries in unique ways. Members of the long-form generative art community have used Deca to curate customised galleries of outputs within and across projects, creating shared digital spaces to enhance collective appreciation of these projects. The entries in Deca’s Zancan x Zeneca gallery competition, in particular @meta_charsiu’s winning gallery, demonstrate how thoughtful curation and commentary can elevate the display of generative art works.

https://twitter.com/0xDecaArt/status/1547644666845073410

At the platform level, Art Blocks and fxhash have also engendered vibrant communities. Using just Art Blocks as an example, in terms of numbers, the cumulative number of unique Ethereum wallets holding at least one Art Blocks NFT stands at around 75,000 as of 1 January 2023, reflecting a more than 50% increase from just a year before, despite an extremely tumultuous 2022 for crypto in general. Art Blocks has also been organising numerous online and offline events for its community, including a 24-hour “Blocktalkathon” on Twitter recently to celebrate its two-year anniversary.

There have also been initiatives by many other organisations to spread the joy about generative art, in particular Bright Moments, which seeks to organise in-person events centred around unique minting experiences for generative art. Kaloh, who publishes an informative newsletter on NFT art, had many positive things to say about Bright Moments’ last event in Mexico City. I have never attended any of these events before, but it is good to see that the impact and reach of long-form generative art extend beyond the digital realm.

Based on my general observations, the long-form generative art community is also open, friendly and highly dedicated. Unlike many other NFT projects, there is a much greater emphasis on the art and the underlying blockchain technology, instead of the financial returns of each project. As Zeneca, an NFT collector and thought leader, has written about the Art Blocks community in 2021 during the last hype cycle:

With Art Blocks, almost everyone comes first for the art and the revolutionary utilization of the Blockchain. You are buying great art first and foremost, and it just so happens that an amazing community has formed around all of it. Creators and collectors come together in an environment that is friendly, informative, inspirational, relaxed. If you’re feeling the FOMO from all the other projects and want to be a part of a killer community, take a step back from the next panda-cat-donkey-cow project and perhaps take a gander at Art Blocks.

Prospects for long-form generative art

To sum up, I believe long-form generative art as a genre can have lasting appeal, and my framework for evaluating this potential value is based on the three “C”s:

1. Content: Do I like the outputs of a particular project?

2. Context: Can I appreciate the creation process of a particular project, and how the different outputs relate to each other and as a whole?

3. Community: Is there a sustainable community with much positive vibes surrounding the project, the artist and/or the platform?

I know art is inherently subjective, and it is futile to try to forecast how the market will shape up in the next few years and decades. But just based on the simple principle that what is appealing will have value, I think long-form generative art is worth exploring, collecting and investing in—in particular, projects with compelling content, rich context and a strong community.

The advent of long-form generative art has also come at a no better time than now. I think we are on the cusp of witnessing several technological shifts converging in the not-too-distant future, in particular, artificial intelligence (AI) and blockchains. As we have seen from the outputs of text-to-image AI-based generative tools (i.e. DALL-E, Midjourney and Stable Diffusion), the marginal cost of content production is going to approach zero. Blockchains will therefore become all the more relevant. We will need NFTs to anchor a semblance of authenticity amidst the deluge of AI-generated content that will descend upon us. And if I may exaggerate a little, NFTs will be critical to preserve our humanity, by ringfencing and securing some portion of art, and the act of creation itself, as a fundamentally human craft.

In this regard, I will hazard a prediction that long-form generative art, as a blockchain-native genre of art, will only increase in cultural relevance, in tandem with NFTs becoming a core building block of our digital lives. Moreover, I think many of today’s projects in long-form generative art will go on to hold tremendous historical value, representing pioneering efforts at a then-nascent stage of this new and dynamic genre of art. Perhaps, the user WhiteRabit1111 puts it the best in an article on Medium:

“There is no limit to the supply of tokenized assets that will emerge in future, but there is an iron-clad hard cap on the scarcity of those in the past.”

On this note, I shall end with this beautiful piece—Meridian #820—the latest entry in my little collection of long-form generative art.

https://opensea.io/assets/ethereum/0xa7d8d9ef8D8Ce8992Df33D8b8CF4Aebabd5bD270/163000820

Disclaimer: Nothing in this essay constitutes financial advice. Please do your own research and be mindful of your own investment objectives before investing into any cryptocurrencies or NFTs.

Cover image by Mathias Reding from Pexels.

If you ask most DeFi participants about which stablecoins they use or have heard of, chances are you will hear a few common names – USDT, USDC, BUSD and DAI – that are currently part of the top four stablecoins by market capitalisation. There is also UST, which was the third largest stablecoin by market capitalisation before its spectacular collapse in May 2022. Sitting below these established names is Frax Finance’s FRAX stablecoin, which has received much less attention even though it is now the fifth largest stablecoin by market capitalisation.

I believe Frax Finance remains an under-the-radar DeFi protocol that bears watching because it is not just trying to build a more well-designed stablecoin (i.e. FRAX), but leveraging it to acquire an outsized influence in the rapidly-growing DeFi space. In fact, in my previous essay on Olympus DAO, I characterised the protocol as a ”community-owned central bank”. But after learning about Frax Finance, I think the latter comes much closer to building DeFi’s version of a central bank, and it is doing this from ground up, starting with its FRAX stablecoin.

(1) The case for a partially-collateralised stablecoin

To understand how FRAX works, it is useful to situate it among the other stablecoins in existence, most of which fall squarely within two camps:

• Fully-collateralised stablecoins: These stablecoins are backed by an equivalent amount of assets or more. In other words, to mint 1 USD worth of such stablecoins, at least 1 USD worth of collateral must be parked in a reserve to back them. Examples of such stablecoins include USDT (or so its creator claims), USDC (which is fully backed by USD cash and short-duration US government treasuries), and DAI (which is in fact overcollateralised, since its collateral includes more volatile crypto assets such as ETH).

• Non-collateralised stablecoins: These stablecoins do not have any collateral to back them, but instead utilise various algorithmic mechanisms to maintain their peg. Terra protocol, whose UST stablecoin was the largest non-collateralised stablecoin by market capitalisation before its collapse, used an associated token, LUNA (now called LUNC), to absorb the price volatility of UST and maintain its peg. When UST fell below 1 USD, Terra incentivised users to burn their UST to mint LUNA, thereby reducing the UST supply and increasing its price back towards 1 USD. Conversely, when UST rose above 1 USD, Terra incentivised users to burn their LUNA to mint UST, increasing the supply and thus reducing the price of UST.

Both types of stablecoins come with their own set of limitations. Fully-collateralised stablecoins have to contend with custodial risk (are its collateral held in a secure place?) and face constraints regarding capital efficiency (their growth is limited by the availability of collateral). On the other hand, non-collateralised stablecoins tend not to inspire much confidence from market participants since they are largely unbacked, and therefore face a significant risk of a bank run that can completely obliterate their value, as shown in the case of Iron Finance’s IRON stablecoin, and most recently, Terra’s UST stablecoin.

Frax Finance took a middle-of-the-road approach with FRAX, by designing it to be a partially-collateralised stablecoin. This does not eradicate the limitations above, but it does moderate them. As a partially-collateralised stablecoin, FRAX is more capital efficient than fully-collateralised or overcollateralised stablecoins – it requires less collateral to maintain its peg. At the same time, by being partially backed (on top of other mechanisms which will be explained in the next section), market participants will feel more assured when using FRAX as compared to other non-collateralised stablecoins, resulting in a lower risk of a bank run.

(2) How does the FRAX stablecoin work?

The name of the FRAX stablecoin is derived from the term “fractional-algorithmic”, which reflects the mechanism by which FRAX maintains its peg to the US dollar, i.e. 1 FRAX = 1 USD.

FRAX is essentially an attempt to incorporate the principles of both fully-collateralised and non-collateralised stablecoins – a part of FRAX’s value is backed by collateral (predominantly USDC for now), while the remaining part is algorithmically stabilised with the help of another token (i.e. FXS, which is Frax Finance’s governance and utility token).

Maintaining the peg

Like other stablecoins, FRAX maintains its peg to USD by allowing for arbitrage via expanding and contracting the supply of FRAX and FXS:

• When the price of each FRAX falls below 1 USD: Users can purchase FRAX on the open market, and then make a profit after redeeming each FRAX for 1 USD of value from the protocol. At the current collateral ratio of around 89%, every FRAX can be redeemed for 0.89 USD of collateral (i.e. USDC) and 0.11 USD worth of newly-minted FXS. In doing so, the supply of FRAX is reduced, and the price of FRAX should trend upwards 1 USD.

• When the price of each FRAX rises above 1 USD: Users can mint new FRAX by depositing 1 USD of value into the protocol, and profit from selling the minted FRAX for over 1 USD each on the open market. At the current collateral ratio of around 89%, the user must deposit 0.89 USDC and 0.11 USD worth of FXS (which will be burned) to mint each new FRAX. In doing so, the supply of FRAX will increase, and the price of FRAX should trend downwards towards 1 USD.

Determining the collateral ratio

What sets FRAX apart from other stablecoins is the fact that it is not only partially collateralised, but that it also allows for a dynamic collateral ratio that is determined by the market. The protocol itself makes no assumption about what this collateral ratio should be, and simply keeps the ratio at whatever the market demands to ensure that each FRAX is priced at 1 USD. In a way, the collateral ratio at any given point in time reflects the extent of the market’s confidence in a non-collateralised stablecoin.

When FRAX was first launched in December 2020, the collateral ratio was set at 100%, and the protocol adjusted it once every hour in steps of 0.25%:

• When the price of each FRAX fell below 1 USD: The protocol increased the collateral ratio by one step per hour, to help restore confidence in FRAX by increasing its backing. This meant that if the price of FRAX was below 1 USD a majority of the time within a certain time period, the collateral ratio would trend upwards towards 100%.

• When the price of each FRAX rose above 1 USD: The protocol decreased the collateral ratio by one step per hour, which would facilitate the minting of new FRAX to bring the price back down towards 1 USD. In other words, if the price of FRAX was above 1 USD a majority of the time within a certain time period, the collateral ratio would trend downwards.

In February 2021, Frax Finance updated its mechanism to determine the collateral ratio, incorporating a metric called the “growth ratio”, which measures the amount of FXS available as liquidity on decentralised exchanges (DEXs) against the total supply of FRAX. Using this growth ratio helps the protocol to ensure that any reduction in the collateral ratio can be supported by the relative supply of FXS and FRAX.

After all, a larger supply of FRAX is harder to stabilise algorithmically. Hence, the protocol should only reduce the collateral ratio when there is sufficient liquidity for FXS (as reflected by a higher growth ratio) to support doing so. Otherwise, in the case of insufficient FXS liquidity (as reflected by a lower growth ratio), reducing the collateral ratio would mean that while users who redeemed their FRAX would get more newly-minted FXS, they would also face greater price slippage when selling these FXS. This may lead to undesirable negative feedback loops, such as a price collapse for FXS over time.

Buybacks and recollateralisation

Due to FRAX’s dynamic collateral ratio, there are periods of time when the protocol has either too much or too little collateral. To correct for such surpluses or deficits, the protocol has built in two swap functions to rebalance the amount of collateral it has: buybacks and recollateralisations.

When the value of the collateral in the protocol dips below the current collateral ratio, a user can call the “recollateralise” function to check whether this is the case. If so, the user will be allowed to add collateral up to the amount needed to reach the target collateral ratio. In return, the user will get newly-minted FXS at a discount rate, which is set at 0.20%. This provides an incentive for arbitragers to recollateralise the protocol whenever there is a deficit in collateral.

On the other hand, when the value of the collateral in the protocol rises above the current collateral ratio, a user can call the “buyback” function to verify whether this is the case. If so, the protocol will use the excess collateral to swap for FXS and then burn it. This reduction in FXS supply will benefit all FXS holders, akin to how share buybacks work in the stock market.

(3) What is the value of holding FXS?

Now that you understand how FRAX works as a stablecoin, it is time to look at how the protocol is able to accrue value for itself, and by extension, FXS holders.

Fundamentally, the value of FXS is tied to the demand and adoption of FRAX:

• If you recall, when demand for FRAX decreases (thereby pushing the price below 1 USD), users are incentivised to redeem FRAX for the underlying collateral and some newly-minted FXS. The supply of FXS will therefore be inflationary if there is poor demand for FRAX.

• In contrast, when demand for FRAX increases (thereby pushing the price above 1 USD), users are incentivised to mint new FRAX by depositing collateral and burning FXS. The supply of FXS will therefore be deflationary if there is strong demand for FRAX.

Enabling demand for FRAX

The key to increasing the value of FXS is therefore to ensure strong demand for FRAX.

There are a few common things that DeFi protocols do to build demand for their tokens, and not surprisingly, Frax Finance has been pursuing these as well:

• Incentivising liquidity: A critical enabling factor of demand for any DeFi protocol’s token is liquidity, without which users will not have the confidence to transact with the token, given the risk of price slippage. This is why DeFi protocols have typically devoted substantial resources to incentivise users to provide liquidity for their tokens. Frax Finance is no different, and it has allocated 60% of FXS’ total supply (capped at 100 million FXS tokens) for liquidity incentives. In other words, if users provide liquidity for FRAX in an eligible liquidity pool on a DEX, they will be able to get additional FXS rewards (see here for the list of active FRAX liquidity pools that receive liquidity incentives in FXS and other tokens). Nevertheless, liquidity incentives are often not sustainable, and should only be used primarily when a protocol has just started out to bootstrap liquidity from nothing. In the case of Frax Finance, the emissions schedule of FXS for liquidity incentives will halve every year based on current parameters. This should help to pare down the dilution in FXS supply from liquidity incentives and resultant selling pressure.

• Partnering other protocols: While critical, liquidity alone will not guarantee demand, especially as liquidity incentives decrease over time. Frax Finance has to build up organic demand for FRAX, and this is best achieved when other DeFi protocols find it useful to transact or hold FRAX. Partnerships with other protocols are therefore something that Frax Finance has been focused on, to develop use cases for FRAX. For instance, Frax Finance and Olympus DAO, which I previously wrote about, partnered to launch a dual-incentivised FRAX-OHM pool on Uniswap, and have undertaken a DAO-to-DAO swap to accumulate each other’s tokens in their respective treasuries, among other collaborations.

• Going multi-chain: Expanding FRAX beyond its native chain, i.e. Ethereum, has also helped to increase the total addressable market for Frax Finance, in terms of users and protocols it can partner with. To date, both FRAX and FXS are deployed on 14 blockchains. Frax Finance has a bridge mechanism to maintain a consistent global state for FRAX and FXS, and ensure that the FRAX peg is kept tight across all blockchains that it is deployed on.

While the market capitalisation of FRAX still pales that of the top four stablecoins, it has been steadily growing since FRAX was launched in December 2020. At the same time, much of the growth in FRAX’s market capitalisation occurred between October 2021 and the start of this year, just before the cooldown in DeFi markets that began in end-January and the much sharper downturn in May following Terra’s collapse.

This suggests that besides Frax Finance’s ongoing demand generating efforts as mentioned above, there may be other factors at play behind that sudden spurt of demand for FRAX in Q4 2021. Here, we turn to another of Frax Finance’s innovations, “Algorithmic Market Operations”, which started taking off around the same time period.

Algorithmic Market Operations (AMOs)

The concept of AMOs was first introduced by Frax Finance as part of its upgrade to Frax v2, announced in March 2021. These AMOs are enacted by smart contracts, which Frax Finance calls “AMO controllers”. They essentially enable Frax Finance to autonomously pursue market operations in ways that can drive greater value to the protocol, without jeopardising the peg and stability of FRAX, which is central to what Frax Finance is about.

As Sam Kazemian, the founder of Frax Finance, explained in his Medium post announcing Frax v2:

Simply put, an AMO controller is an autonomous contract that enacts arbitrary FRAX monetary policy so long as it does not lower the collateral ratio and change the FRAX price. This means that AMO controllers can perform open market operations algorithmically (that’s where they get their name), but they cannot simply mint FRAX out of thin air and break the peg. This keeps FRAX’s base layer stability mechanism pure and untouched which has been the core of what makes our protocol special and has inspired other smaller projects. It also keeps AMO controllers economically disciplined knowing that there is a ceiling for the amount of FRAX that can be unbacked which stops the protocol from unrestrained money printing of classical central banks.

Each AMO can be seen as a module that operates on top of Frax Finance’s base “fractional-algorithmic” mechanism. While each AMO can run different market operations and thus achieve different objectives for the protocol, they all share the following four functions which builds on the base mechanism:

• Equilibrium: Under the base mechanism, the protocol will not do anything when it is at equilibrium, i.e. the price of FRAX is at its peg of 1 USD. For an AMO, this is the primary space in which it will run its market operations.

• Decollateralise: If the price of FRAX exceeds 1 USD, the base mechanism will aim to increase the supply of FRAX by essentially reducing the collateral ratio to encourage the minting of new FRAX. An AMO likewise has to include a similar “decollateralise” function to reduce the collateral ratio and help bring the protocol backs towards equilibrium when the price of FRAX exceeds 1 USD.

• Recollateralise: Conversely, if the price of FRAX falls below 1 USD, the base mechanism will aim to increase confidence and demand for FRAX by essentially increasing the collateral ratio. An AMO likewise has to include a similar “recollateralise” function to increase the collateral ratio and help bring the protocol backs towards equilibrium when the price of FRAX goes below 1 USD.

• FXS1559: This is a function, named in homage to EIP1559, that builds on the “buyback” function in the base mechanism and ensures that every AMO is able to accrue value to FXS when the protocol has excess collateral. The FXS1559 function can calculate the value of the excess collateral above the protocol’s collateral ratio at a given time interval, and mint an equivalent amount of new FRAX at the collateral ratio. It then uses the newly-minted FRAX to buy FXS from the open market and burn it, thus reducing the supply of FXS and accruing value to FXS holders. Unlike the “buyback” function in the base mechanism which uses the excess collateral to directly swap for FXS and burn it, the FXS1559 function has the added step of using the excess collateral to mint new FRAX first, which will help to also expand the supply of FRAX.

Frax Finance has designed Frax v2 to incorporate multiple AMO controllers, with anyone able to propose, build and deploy AMOs, subject to the protocol’s governance process. Currently, the protocol has several AMOs in operation, with the key ones outlined below:

• Collateral Investor AMO: This AMO enables the protocol to tap on its idle collateral (currently only USDC) and move them to selected DeFi protocols (e.g. Aave, Compound and Yearn) to generate yield. Collateral that is invested with the option for immediate withdrawals do not count as lowering the collateral ratio. Revenue generated from the yield that places the value of the protocol’s collateral above the collateral ratio can be used to mint FRAX and burn FXS via the FXS1559 function to accrue value to FXS holders.

• Curve AMO: This AMO allows the protocol to deploy its idle USDC collateral or mint new FRAX (to the extent that it will not change the collateral ratio) into its FRAX3CRV liquidity pool on Curve. This helps to increase liquidity for FRAX and tighten its peg, while simultaneously allowing the protocol to earn revenue and CRV rewards from providing liquidity on Curve.

• Lending AMO: This AMO enables the protocol to mint FRAX directly to lending/borrowing protocols such as Aave and Compound, from which users can obtain FRAX by borrowing, instead of minting it via the base mechanism, i.e. by depositing collateral and FXS. The protocol earns revenue from the interest paid by borrowers. In addition, as lending/borrowing protocols require borrowers to overcollateralise their positions, the operations of the lending AMO do not lower the collateral ratio. This effectively allows the protocol to mint new FRAX at will as long as FRAX does not lose its peg. Given this low cost of money creation, Frax Finance claims that it can offer more competitive interest rates for lending FRAX as compared to other stablecoins. Ultimately, the lending AMO allows Frax Finance to influence the interest rate for FRAX by minting or removing FRAX accordingly – a powerful lever that allows it to retain significant influence over market demand for FRAX.

What is noteworthy about the impact of these AMOs was that since October 2021, the large amount of new FRAX that were minted seemed to have been through the AMOs, rather than the base mechanism. According to the FXS dashboard on Frax Finance’s website, there have not been much FXS burned since October 2021.

This seems counter-intuitive. After all, even if much of the new FRAX was minted by the AMOs (in particular the Curve AMO), shouldn’t the FXS1559 function work to buyback and burn FXS using the profits from the AMOs? This was how the AMOs were supposed to accrue value to FXS holders.

Staking FXS

It turns out that the FXS1559 function is not the only way to do so. In fact, Frax Finance has also directly channelled its AMO profits to FXS holders who stake their FXS. Initially, 50% of the profit from the AMOs (i.e. yield that forms the protocol’s excess collateral) was used to buy and burn FXS via the FXS1559 function, while the other 50% was used to buy FXS and distribute them to those who stake their FXS as veFXS.

However, following a governance vote in October 2021, 100% of AMO profits have since been channelled to the latter, i.e. veFXS stakers. This was why there was not much burning of FXS since then. As outlined in the governance proposal, this move was to encourage the long-term holding of FXS, by increasing the yield from staking FXS as veFXS.

veFXS stands for vote-escrowed FXS. It is based off the vesting system pioneered by Curve Finance for its CRV token, which can be locked as veCRV. Essentially, users can lock up their FXS tokens and receive veFXS tokens in return, at a rate that depends on the duration of lock-up. Those that stake their FXS for the maximum time period of four years will receive 4 veFXS for every FXS locked, while those who do so for shorter time periods will receive fewer veFXS proportionally. The veFXS tokens are non-transferrable and a user’s veFXS balance will vest linearly, decreasing to 1 veFXS per 1 FXS at zero lock time remaining.

The intent of such a vesting system is to encourage long-term holding of FXS, since users receive more veFXS the longer they lock their tokens for. The utility of holding veFXS, to encourage users to even want to lock up their FXS in the first place, can be summed up by the following:

• Greater voting power: Each veFXS gives the user one vote in governance proposals. If a user wants to have more say in the governance of Frax Finance, then he or she needs to acquire more FXS and lock them for the maximum time period of four years – a long time in the world of crypto.

• Boost in liquidity incentives: Holding veFXS also gives users a greater share of the protocol’s liquidity incentives if they provide liquidity to eligible liquidity pools. veFXS holders also have the ability to vote on where the protocol’s liquidity incentives go through the gauge system, and this can be valuable because users can vote to direct more incentives towards the liquidity pools that they are participating in.

• Yield: Holding veFXS entitles users to receive the protocol’s AMO profits in the form of additional FXS. At the moment, 100% of the AMO profits are distributed to veFXS holders. The annual yield works out to an annual percentage rate (APR) of around 8% at the time of writing, but it can range from about 3% to as high as 20% depending on market conditions, as Sam shared on this Bankless podcast.

Valuing FXS

Having gone through the main ways in which Frax Finance is able to accrue value to FXS holders, the next important question for any potential FXS holder is how should one value FXS?

Since Frax Finance is a revenue-generating protocol (primarily through its AMOs), one way is to use the ratio of FXS’ market capitalisation and the protocol’s annualised earnings to get what is essentially the price-to-earnings (PE) ratio for the protocol. This ratio can serve as a rough measure of Frax Finance’s valuation when compared with other revenue-generating entities.

Given the rather significant price movements over the past month following Terra’s collapse, I thought to calculate Frax Finance’s PE ratio before and after that unfortunate incident, using two arbitrary dates: 1 May and 28 May.

Frax Finance’s trailing PE ratio (based on earnings in the past 12 months) is around 18.96 on 1 May 2022 and 11.80 on 28 May 2022. This is actually significantly lower than the trailing PE ratio for the S&P 500 and Nasdaq 100 as reflected below, but bear in mind that this is an apples-to-oranges comparison. In fact, if I were to narrow down the comparison to real-world companies that are most similar to Frax Finance, i.e. traditional banks, the protocol’s valuation is still higher.

In any case, comparing against other DeFi protocols will be more instructive, but likewise, the limitation is that each protocol is designed differently and there nothing quite like Frax Finance and its AMOs on the market now. A decent comparison could be with MakerDAO, which is a decentralised autonomous organisation (DAO) managing the DAI stablecoin, and earns revenue from fees paid by users for borrowing DAI. Between May 2021 and April 2022, MakerDAO reported a net protocol income of USD 130.9 million. As of 1 May 2022, its market capitalisation was USD 1.31 billion based on its MKR token, which translated to a trailing PE ratio of around 10.01. As of 28 May 2022, assuming MakerDAO’s profits in the past 12 months remain constant, its trailing PE ratio based on the market capitalisation of MKR then would be around 8.01.

From this, you can see that Frax Finance has a more demanding valuation than MakerDAO. That said, another perspective is that Frax Finance probably has a higher growth potential, given that the partially-collateralised FRAX is more capital-efficient than the over-collateralised DAI. Furthermore, the market capitalisation of FRAX is still only about one-quarter that of DAI – if the supply of FRAX continues to grow, chances are that much more value can further accrue to Frax Finance.

Of course, there are many more ways to assess the valuation of a potential investment, and to do so more rigorously. But my primary aim to use the PE ratios was to simply show that the valuation of Frax Finance is not crazy. FXS could even be considered under-valued when compared against growth stocks, given that the current trailing PE ratio for the Nasdaq 100 is at 25.43. Moreover, unlike many of these companies which are not even profitable in the first place, Frax Finance already has a sustainable source of income through its AMOs that it can further grow as it works to expand the use of FRAX across DeFi.

(4) What is the end-game for Frax Finance?

Ultimately, investing is more than just assessing financial statements and valuations. It is also about making calculated (and optimistic) bets on the future, which often may not have a quantifiable basis. So the most salient question for me when looking at Frax Finance is more than just whether it is fairly-valued today, but whether it has a compelling and achievable long-term vision – an end-game that it can works towards and bring into reality.

Achieving the “holy trinity”

Sam has been clear about Frax Finance’s end-game – to achieve what he calls the “holy trinity” of DeFi, in which a protocol is able to have its own stablecoin, liquidity ecosystem, as well as lending market. This essentially allows the protocol to not only issue its own currency (its stablecoin), but also control the supply and demand of its currency (via its liquidity ecosystem and lending market). This is very much like how actual central banks work, which buy and sell securities on the open market to influence money supply in their respective jurisdictions.

https://twitter.com/Rewkang/status/1479193841680928768

Frax Finance has been building towards this “holy trinity” systematically:

• Stablecoin: Frax Finance’s core product is its FRAX stablecoin, which has done its job remarkably well – it has never gone off its peg since its launch, even during the recent turmoil caused by Terra’s collapse. The distinctive feature of FRAX is not only that it is partially collateralised and partially algorithmically-stabilised, but that its collateral ratio is also dynamic. This adds what Sam has called “an anti-bank run mechanism” to FRAX, in that if market confidence in FRAX falls for whatever reason, the collateral ratio will increase to help restore confidence in FRAX. All in all, it is clear that a lot of thought has gone into the design of FRAX to ensure that it is fundamentally sound.

• Liquidity ecosystem: Frax Finance has built a specialised automated market maker (AMM) called Fraxswap. Unlike generalised AMMs or DEXs like Uniswap or Curve, Fraxswap is envisioned to be targeted at DAOs. Its aim is to allow these DAOs to rebalance large amounts of tokens efficiently via a mechanism called the time-weighted average market maker (TWAMM), which breaks a large order into “infinitely many infinitely small sub-orders” to ensure smooth execution over time with minimal price slippage and gas costs. Ultimately, by creating and controlling its own liquidity ecosystem, Frax Finance can expand the reach of FRAX as a source of liquidity for many other assets.

https://twitter.com/GoodPepe_/status/1505369081054384129

• Lending market: Frax Finance has also indicated that it is working to build its own lending market called Fraxlend. This would be similar to the Lending AMO, except that instead of minting new FRAX on other lending/borrowing protocols for users to loan, Frax Finance can do this on its own platform, i.e. Fraxlend. In doing so, Frax Finance will be able to better control its own interest rates, which will ensure that it retains significant influence over market demand for FRAX. At the same time, Fraxlend may also signal a shift in FRAX’s model – from being a stablecoin that is only partially backed by collateral, to one that is also backed by credit. This would strengthen the resilience of FRAX, as the portion of FRAX that is algorithmically stabilised by FXS would then also be backed by on-chain loans, reducing the reflexive risk that had brought down UST and LUNA.

https://twitter.com/davesaidthat_/status/1525013488149549057

Besides working towards the “holy trinity”, it is also worth noting that Frax Finance is building another innovative product – the Frax Price Index (FPI) stablecoin whose value tracks the 12-month Consumer Price Index in the US. In other words, FPI is designed to be an inflation-adjusted stablecoin that can serve as a true store of value. It is meant to track inflation through the yield generated from the FRAX used to mint FPI. As the detailed documentation for FPI is still being worked on, the exact mechanisms by which this yield would be generated is not yet clear.

https://twitter.com/AvgJoesCrypto/status/1514622665243471878

Nevertheless, if Frax Finance can ensure that FPI is able to generate sustainable yield to track inflation over the long term, it would serve as yet another way to entrench the protocol across various use cases in DeFi: transactions (with FRAX), trading (with Fraxswap), borrowing/lending (with Fraxlend), and capital preservation (with FPI).

Risks

Of course, like virtually all projects in the nascent and largely experimental DeFi space, Frax Finance still faces significant risks. These would include:

• Smart contract risk. Like other DeFi protocols, Frax Finance is built on smart contracts. Any flaw or loopholes in their code could therefore lead to exploits or hacks, resulting in compromised or stolen assets. While Frax Finance’s smart contracts are not immune to this risk, the fact that they have been audited by third-parties should provide some assurance. That said, as its latest audit report by Trail of Bits in December 2021 stated, “every exchange or sidechain that the protocol interacts with adds to the attack surface”. With Frax Finance linked to many other DeFi protocols through its AMOs, it is also exposed to the smart contract risk of these protocols.

• Custodial and regulatory risk. The entirety of FRAX’s collateral is currently in one centralised stablecoin, i.e. USDC. This means that if USDC fails, the value of FRAX would be at risk. Likewise, if Centre (the joint venture between Coinbase and Circle to create the USDC stablecoin) is forced by regulators to block Frax Finance from using USDC, FRAX will also be impacted. Nevertheless, Sam has clarified that most of FRAX’s supply lies in smart contracts in other DeFi protocols due to its AMOs; hence, regulators are unable to blacklist FRAX fully unless they blacklist these other protocols that use FRAX. He added that Centre would also have no incentive to allow this to happen, as it would effectively destroy most of USDC’s use cases.

• Execution risk. Much of Frax Finance’s potential depends on Sam and his team’s ability to continue executing their vision. If anything happens to them or their ability to work on the protocol, there is a chance that the protocol may not live up to its potential. Similarly, like central banks in the real world, the success of Frax Finance will also hinge on sound monetary policy – its AMOs have to work, if not the protocol will not be able to expand the supply of FRAX and accrue value to FXS holders. This will certainly not be easy, given the nascence and volatility of crypto markets as we have seen this year, especially since Terra’s collapse. There will certainly be more headwinds in the future, and Sam and his team will have their work cut out as they steer Frax Finance through these challenges.

At the end of the day, as Sam has acknowledged, Frax Finance is “highly experimental and risky”. There is every chance that the protocol can fail, from the risks stated above and beyond. With that in mind, it is imperative that one should not be going all in on FRAX or FXS. Even if the potential rewards seem attractive, always be sure to diversify your investments. After all, nothing is ever certain in this world.

(5) Concluding thoughts

I was only able to write this essay in fits and starts over the course of more than two months due to my full-time work commitments. During this time, I have witnessed both ends of the hope-fear spectrum regarding Frax Finance:

• The former occurred when Terra partnered with Frax Finance to launch a new stablecoin liquidity pool on Curve, called the “4pool” comprising of UST, FRAX, USDC and USDT. Given that UST and FRAX were the leading non-collateralised algorithmic stablecoins in the market then, many expected that their partnership would help anchor market confidence in both stablecoins, and push the adoption of algorithmic stablecoins further. If you look at the price of FXS, it spiked significantly when the 4pool was announced in early April.

• Unfortunately, the latter occurred shortly after. No long after the 4pool launched, UST failed. Its collapse in early May predictably impacted perceptions of FRAX, since both were algorithmic stablecoins. As a result, the price of FXS has fallen significantly since then.

That said, regardless of market sentiments, what I have seen from Sam and his team have been a steadfast commitment to continue building, and to make the protocol as solid as it can be. In fact, Sam has written that he wants “FRAX to be as safe as the best”, and that “just because FRAX is and has been at peg always doesn’t mean we can’t make it better”. As a holder of a small amount of FXS, this growth mindset is reassuring to me.

Protocols that will survive the current bear market will be those that are built to last. From what I can see, Frax Finance fits into this mold. While it is impossible to predict the movement of markets over the short term, I still remain optimistic about the value proposition of DeFi, and Frax Finance’s potential to occupy a strong position in this nascent yet fast-evolving space over the long term. The design of FRAX seems fundamentally sound, and the protocol’s AMOs appear to be well-placed to expand the supply of FRAX and generate revenue for FXS holders sustainably.

My hope is that Sam and his team can continue building on their good work so far, and see through their vision of creating DeFi’s equivalent of a modern central bank – one that will keep on chugging away to make sure that our future on-chain economies remain stable as much as possible, with ample liquidity and sustainable credit facilities.

Disclaimer: Nothing in this essay constitutes financial advice. Please do your own research and be mindful of your own investment objectives before investing into any cryptocurrencies, DeFi protocols or NFTs.

Cover image by Pixabay from Pexels.

Since its official launch earlier this year (on 23 March to be precise), it has grown to become not only one of the most significant DeFi protocols, but also one that has been highly polarising. Its most ardent supporters have spawned a canon of memes to promote Olympus DAO’s work, while critics have slammed the entire initiative as a Ponzi scheme.

I believe that cutting through all this noise will be crucial for anyone interested to get into DeFi. Whether you believe in Olympus’ mission or not, it is important to at least understand how it works, for it is becoming a foundational platform connecting many other DeFi protocols.

(1) What is Olympus DAO’s objective?

Olympus DAO is a decentralised autonomous organisation that operates a DeFi protocol. As its website clearly proclaims. it is seeking to build a “decentralised reserve currency”, which takes the form of its own token called OHM.

Before we go into the “tokenomics” of OHM, let’s unpack what Olympus DAO means by a “decentralised reserve currency”:

• First, a currency that is truly decentralised cannot be controlled by a single entity. This means that it should not be directly linked to fiat currencies, which are exclusively within the control of each country’s central bank.

• Second, a reserve currency has to have two things: (i) deep liquidity so that users can get it when they need it; and (ii) some form of backing. This backing can be in the form of a tangible asset such as gold (for the US dollar when it could be exchanged for gold before 1971), or faith that the US government will repay its Treasury securities (for the US dollar currently).

Combining these together, the “decentralised reserve currency” that Olympus DAO envisions, i.e. OHM, should therefore have the following characteristics:

• OHM should ideally be backed by assets that are not directly linked to fiat currencies. Unlike the US dollar, OHM has to be backed by something since there is no way a DeFi protocol can command as much as faith as the US Government now.

• OHM should also be extremely liquid, so that people can easily use it as the foundation of their activities across DeFi, e.g. to provide liquidity or swap for other crypto assets.

(2) Why is there a need for a decentralised reserve currency in the first place?

Now, you might be wondering why there is a need another decentralised currency, when we already have established cryptocurrencies like Bitcoin and Ethereum, which are already decentralised.

The answer is that these cryptocurrencies are highly volatile. One calculation of Bitcoin’s historical volatility puts it as around 2.4 times more volatile than the S&P 500, and 5.5 times more volatile than the Euro (against the US dollar) over a 200-day period. This lack of price stability means that you probably would not want to use Bitcoin, or most cryptocurrencies for that matter, as a store of value.

Hence, for all the talk about the value of cryptocurrencies as decentralised assets that can serve as a hedge against inflation in fiat currencies, much of DeFi has become underpinned by US dollar-pegged stablecoins such as USDT and USDC. As Zeus, the pseudonymous founder of Olympus DAO, wrote in the protocol’s first post on Medium, “there’s a strange irony to the fact that the most utilized cryptocurrency is really just a digitized dollar.” Nevertheless, users clearly want some semblance of stability when engaging in DeFi.

Olympus DAO, however, thinks it can do better than the current crop of stablecoins.

While stablecoins may achieve a stable fiat value, Olympus DAO believes that they are not ideal as a reserve currency because their real value or purchasing power may not necessarily be stable. By virtue of being pegged to fiat currencies, stablecoins are likewise at risk of their real value being eroded by the inflationary policies of governments.

Stablecoins also have issues related to centralisation that a decentralised reserve currency can avoid:

• Most of the stablecoins used today are issued by centralised entities, e.g. Tether for USDT and Centre for USDC. This creates a single point of failure for these stablecoins. If their issuing companies are mismanaged and go bust, the stablecoins may become worthless (you cannot go back to them to change your stablecoins back to fiat).

• The companies can also act unilaterally to prevent you from using your stablecoins. They have blacklisted certain wallet addresses in the past, when faced with demands from regulatory bodies. This vulnerability to regulatory action may not be unwarranted, especially in the case of Tether, which has attracted much controversy over the backing of USDT. Nonetheless, there is a narrative mismatch here—if DeFi’s value proposition is anchored on it being a decentralised alternative to the traditional financial system, then it does not gel well to have much of its liquidity denominated in centralised stablecoins, which can be controlled by the issuing companies themselves or by governments.

• Some decentralised stablecoins also had to rely on centralised stablecoins (to some extent) in order to maintain their peg. In particular, MakerDAO’s DAI, which is the first “decentralised” stablecoin to gain widespread use, has been partially collateralised by USDC since March 2020, when the cryptocurrency markets crashed due to uncertainty over the impact of COVID-19. Other centralised stablecoins, e.g. USDT, USDP and GUSD, were also added as collateral assets later that year. Despite its attempt to become a decentralised stablecoin, DAI remains indirectly linked to centralised entities.

Instead of risking depreciation and indirect centralisation by relying solely on stablecoins, Olympus DAO wants OHM to be an independently-valued digital currency, which can serve as a better alternative for the US dollar as a store of value in DeFi. In Zeus’ own words:

OHM will be most successful if it is positioned as a USD-alternative primitive for DeFi. We will work hard to get OHM integrated into money markets, DEX pairs, and yield farming protocols. We will also encourage its use as an alternative to holding USD. Eventually, you’ll be able to quote prices in OHM with a reasonable expectation that they’ll remain the same in the future. At that point, we hope to see people flee to OHM as a safe haven during risk-off periods rather than USD. We will do as much as we can to realize this, but ultimately it is a mindset and belief system (like any money or currency) that will rely on the support of you, the community.

(3) How does OHM work?

Backed, not pegged

As mentioned earlier, a reserve currency ought to be backed by something. For OHM, Olympus DAO decided at its inception that 1 OHM has to be backed by at least 1 DAI in its treasury. (Despite not being completely decentralised, DAI was chosen because it was familiar, and the team wanted to minimise complexity at the initial stage.)

However, unlike DAI itself or other stablecoins, OHM is not pegged to the US dollar. The price of OHM is meant to be free-floating, i.e. it is determined by the market. In other words, 1 OHM can easily trade at a price below or above 1 DAI (or 1 USD assuming DAI does not lose its peg).

At the same time, Olympus DAO’s protocol controls the supply of OHM algorithmically based on the price of OHM:

• When OHM trades below 1 DAI, the protocol will automatically buy back OHM (it can do so because the treasury backs every OHM with at least 1 DAI). By burning this OHM, the protocol can therefore reduce the total supply of OHM and increase the price of each remaining OHM until it goes back to 1 DAI. This process makes it certain that OHM will not trade below its intrinsic value of 1 DAI over the long term.

• When OHM trades above 1 DAI, the protocol can mint and sell new OHM. This will increase the supply of OHM and drive down its price towards 1 DAI.

From the price history of OHM since Olympus’ inception, you can see that OHM has traded much higher than this 1 DAI / 1 USD price floor. Thus far, OHM’s all-time high price was around USD 1,415 (25 April 2021), and its all-time low price was around USD 162 (23 May 2021).

Now, you might wonder why people have been willing to pay such a hefty premium for OHM given that its effective price floor is actually only 1 USD?

The short answer is that Olympus DAO is currently still in its initial stage—it is prioritising growth over price stability, and seeking to accumulate more assets and liquidity in its treasury. The market is therefore willing to pay a premium on OHM now, to capture this potential growth in the value controlled by Olympus DAO, which in turn will be shared with those who hold and stake OHM. (Before we discuss how Olympus DAO does this, we will make a slight detour first to discuss liquidity, which provides a foundation for Olympus DAO to grow its protocol-controlled value, while protecting OHM holders.)

Protocol-owned liquidity

Liquidity refers to the ease with which an asset can be converted into another asset without affecting its current market price. As mentioned earlier, this is another important feature of a reserve currency. After all, if you want your currency to be the currency that people fall back upon as a safe store of value, then they must be able to easily convert their assets into your currency without affecting its price.

In DeFi, liquidity is something that protocols have been fighting hard over. The fate of a protocol’s tokens, and by extension its value, often lies in whether they have sufficient liquidity. This is why many DeFi protocols have typically offered substantial incentives for people to provide liquidity, resulting in the high yields that DeFi is known for.

However, this method of obtaining liquidity is fundamentally unsustainable. Nothing stops third-party liquidity providers from withdrawing their liquidity when the next hyped project with higher yields appear, or when the overall crypto market turn bearish and they want to convert their liquidity into safer assets. The result is that protocols have to effectively keep paying high “rents” for their liquidity, which imposes a perpetual cost on them.

Olympus DAO seeks to overcome this limitation by seeking to own the liquidity of its own tokens. It is one of a newer generation of Defi protocols—often dubbed “DeFi 2.0”—that have pioneered the concept of “protocol-owned liquidity”. Instead of continuously rewarding third-parties to provide liquidity for its OHM tokens, Olympus DAO structures its incentives to instead acquire this liquidity from third-parties outright, and thus permanently own it.

Bonding

The mechanism by which Olympus DAO acquires liquidity for OHM is referred to as bonding.

But before we discuss how bonding works, we must first understand how liquidity is actually provided within DeFi:

• Liquidity is most often supplied in what are called decentralised exchanges (DEXes). These DEXes are the functional equivalents of a crypto exchange like Coinbase, except that they allow users to exchange crypto assets without the need of any third-party.

• This is possible through an automated market making mechanism. Here, instead of users trading a pair of assets with each other as buyers and sellers (e.g. exchanging DAI for OHM, and vice versa), they do so against a liquidity pool containing both assets.

• Generally speaking, the price of both assets within the liquidity pool is determined by a constant product formula, whereby the product of the quantities of the first asset (x) and the second asset (y) must always be a constant (x * y = k). So if I buy OHM from the OHM-DAI liquidity pool, I am effectively adding DAI to the pool and taking out OHM. As the supply of OHM within the pool falls, the OHM-DAI exchange rate will have to shift in favour of OHM (i.e. the price of OHM increases in terms of DAI) so that k remains constant.

• Due to this constant product formula, the liquidity pool of a pair of assets can maintain a state of equilibrium, with the total value of each asset in the pool always equal. What changes is the quantity of the assets within the pool, as well as the exchange rate between them, i.e. their prices. (You can refer to the screenshot below for a worked example from a DeFi course I took on Coursera.)

• In addition, should the prices of the assets within a liquidity pool in a DEX deviate from the market prices on other exchanges, traders can take advantage by arbitraging the liquidity pool until the prices of the assets are back in line with market prices.

• For the purposes of understanding how Olympus DAO’s bonding mechanism works, what you should know about DEXes is that anyone can actually supply their assets to a liquidity pool, and earn a cut of the transaction fees. There is no minimum amount in terms of the liquidity that each person can provide. The only constraint is that you have to supply an equal amount in value for both assets into the pool.

• For instance, if I have USD 2,000 worth of OHM to spare and I want to supply liquidity to the OHM-DAI pool, I will first have to convert USD 1,000 of my OHM into DAI, and then supply USD 1,000 of both OHM and DAI into the pool. In return, I will get a liquidity pool token from the DEX, which signifies my share of the assets within the liquidity pool.

Such liquidity pool tokens for selected pools containing OHM tokens are what Olympus DAO seeks to acquire through bonding. By owning these tokens, Olympus DAO will own the underlying assets within the liquidity pool. It can therefore ensure that those assets remain in the pool, and guarantee liquidity in perpetuity for those holding OHM tokens.

To incentivise owners to transfer their liquidity pool tokens over, Olympus DAO offers them a deal:

• They can “bond” their liquidity pool tokens with Olympus DAO, and in return, obtain OHM tokens at a discount to the current market price.

• Using my earlier example, if my liquidity pool tokens for the OHM-DAI pool are worth USD 2,000, I can bond them to get discounted OHM. If the bond discount is 5% and the current price of each OHM is USD 100, I will be then able to obtain a total of around 21.05 OHM (at the bond price of USD 95 per OHM).

• That said, these discounted OHM will not be released to me all at once. Instead, they vest linearly over time (currently set at around 5 days), so that bonders do not sell all their discounted OHM at once for an immediate profit.

To understand how the bond discount is derived, we must first be aware that the price of bonds is actually set independently from the market price of OHM. The documentation on its website outlines the equations Olympus DAO relies on to derive the bond price, but how it works is as follows, in what I hope is more intuitive language:

• When someone bonds their liquidity pool tokens, Olympus DAO likewise receives the bonded liquidity pool tokens over the five-day vesting period (just as the person receives the discounted OHM over the vesting period).

• The value of the bonded liquidity pool tokens that are awaiting entry into Olympus DAO’s treasury (in USD) is divided by the total supply of OHM to give a figure called the “debt ratio”. This is a measure of the total outstanding debt borne by Olympus DAO at the moment.

• This “debt ratio” is then multiplied by a scaling variable called the “Bond Control Variable” (BCV) to derive the “bond premium”. The bond price is the sum of 1 USD (the intrinsic value of each OHM) and the bond premium (i.e. bond price = 1 + debt ratio * BCV).

• As more of the bonded liquidity pool tokens enter Olympus DAO’s treasury over the vesting period, the debt ratio and hence the bond premium will decrease accordingly, thereby reducing the bond price.

• On the whole, bond prices will increase whenever people bond their liquidity pool tokens with Olympus DAO. A large enough bond purchase may even increase the bond price such that it exceeds the current market price, resulting in a negative discount. Nevertheless, as the bonds vest, the bond price should go then down, and the bond discount should eventually become positive again.

• This dynamic ultimately ensures that Olympus DAO can take on bonded liquidity pool tokens sustainably without incurring too much debt, while also maintaining market demand for OHM through the bond discount.

• In addition, by adjusting the BCV, Olympus DAO can calibrate the amount of liquidity it plans to acquire. For example, if more liquidity in the OHM-DAI pool is desired, Olympus DAO can reduce the BCV for the OHM-DAI liquidity pool tokens. This reduces the rate at which the bond price increases, and makes it more attractive for people to bond their OHM-DAI liquidity pool tokens. (The current BCV values for the different liquidity pool token and asset that can be bonded with Olympus DAO are shown on this dashboard.)

Bonding has thus far been an effective mechanism for Olympus DAO to acquire liquidity for OHM, and thus increase its protocol-owned liquidity. For instance, Olympus DAO now owns more than 99% of the liquidity for the OHM-DAI pool on SushiSwap as well as the OHM-FRAX pool on Uniswap V2.

There are also material benefits to owning one’s liquidity too. The transaction fees for swaps, which is 0.3% of the trading amount on Uniswap V2 and SushiSwap, mainly accrue to the liquidity pool owners, and is split according to their respective stake. By controlling almost all of the liquidity for OHM, Olympus DAO is able to capture revenue from anyone who swaps their OHM for other tokens.

Furthermore, bonding is not only used to enable Olympus DAO to grow its protocol-owned liquidity.

Olympus DAO also uses the same bonding mechanism to acquire assets for its treasury, thus increasing its protocol-controlled value. For users, this means that besides bonding liquidity pool tokens, they can also bond certain assets directly in exchange for discounted OHM.

In its early days, Olympus DAO started with bonding DAI (on top of bonding OHM-DAI liquidity pool tokens). Since then, it has accumulated a diverse range of decentralised assets in its treasury through bonding—this includes other decentralised stablecoins, i.e. FRAX (which is partially collateralised with USDC), LUSD and UST, as well as other non-pegged assets, such as ETH and SUSHI. (See the governance proposal, OIP-51, for an initial list of tokens that Olympus DAO intends to acquire for their strategic value.)

The total market value of all the assets in Olympus DAO’s treasury now stands at around USD 700 million at the time of writing. This is no mean feat given that Olympus DAO began with only 68,260 DAI when it started operating on 23 March 2021—a growth of more than 10,000 times in less than a year.

Like how owning liquidity allows it to obtain the transaction fees paid, Olympus DAO can also generate yield on the assets in its treasury. This is in fact what Olympus DAO has been doing. Its community has recently approved the use of up to 50% of Olympus’ excess reserves (i.e. treasury assets which are not used to back the OHM tokens in circulation) for yield generation in whitelisted DeFi protocols, such as Aave, Compound, and Convex Finance (see OIP-54).

Having explained how Olympus DAO is able to both acquire its own liquidity and grow its treasury through bonding, we now come back to the earlier question: why would anyone want to pay more than the intrinsic value (i.e. 1 DAI) for each OHM token, be it through bonding or trading for it in the open market?

I mentioned that the short answer is that buyers of the OHM token are willing to pay a premium to capture a part of Olympus DAO’s future growth. The long answer requires an explanation of the second key mechanism that Olympus DAO operates, called staking.

Staking

Staking is a mechanism that builds on bonding.

If you recall, the bond price is effectively the sum of the intrinsic value of OHM (i.e. 1 USD) and the bond premium. This premium is the profit earned by Olympus DAO because it is only required to back each circulating OHM with at least 1 DAI (although this can be changed in the future).

90% of this profit from bonding (as well as yields generated from the treasury’s existing liquidity and assets) is then distributed to those who stake OHM, with the remaining 10% going to Olympus DAO’s treasury. How this is done is that Olympus DAO will mint new OHM tokens (backed by the new profits which will stay in the treasury), and these new OHM tokens become the actual reward that will accrue to OHM stakers.

From the user’s perspective, this is how staking works:

• To stake OHM, users have to lock their OHM tokens within the staking contract in Olympus DAO’s application.

• They will then receive an equal amount of sOHM tokens, which effectively reflects their staked OHM. (You can think of sOHM as a token that entities you to redeem the OHM tokens you have deposited in the staking contract at a 1:1 ratio.)

• When Olympus DAO earns a profit and mints new OHM to distribute to stakers, it sends the new OHM tokens to the staking contract.

• Given that there is now more OHM deposited in the staking contract than the sOHM minted by it, the supply of sOHM needs to increase by the corresponding amount. This is what Olympus DAO calls a “rebase”.

• For example, let’s assume there are 1 million OHM staked and thus 1 million sOHM outstanding. If Olympus DAO earns USD 5,000 in profits, which it uses to mint and back 5,000 new OHM, it will send those OHM to the staking contract. There will therefore be 1,005,000 OHM staked and 1,000,000 sOHM outstanding, and the sOHM supply needs to increase by 5,000 or 0.5% to return to balance.

• This increase in sOHM supply is allocated proportionally to how much sOHM you have, i.e. everyone receives the same percentage profit for each rebase. Using the above example, if you have 1,000 sOHM, you will have 0.5% more sOHM after the rebase (i.e. 1,050 sOHM).

The distribution of profits, as manifested in the rebase rewards, is done every eight hours or three times per day. This is automatic, so stakers do not have to do anything except to remain staked. With each rebase every eight hours, the number of sOHM they hold will steadily increase.

Given that the rebases compound, the annual percentage yield (APY) for staking OHM can hit rather astronomical figures. Using the illustrative rebase rate of 0.5% per eight-hour period, 1 sOHM will become approximately 234.42 sOHM after a year—a more than 20,000% increase! This demonstrates the sheer power of compound interest.

In fact, when Olympus DAO first started, its APY was a six-figure number. However, that has since dropped to around 5,000% APY at the time of writing, which is still a pretty insane yield when compared to what you get in traditional finance.

This high APY is likely the reason why many people have been so eager to pay a hefty premium to buy OHM—to pay now for the potential to accumulate much more OHM in the future through staking, as Olympus DAO grows further.

(4) Are the high yields from staking OHM sustainable?

But you might be wondering: Olympus DAO’s APY, even at around 5,000% today, sounds too good to be true—is it a scam?

APY is not everything

The APY figure by itself is actually a distraction because the rebase rate is expected to decrease as Olympus DAO grows. You should therefore not be looking only at the current APY figure to estimate your return on staking OHM over the long term.

In fact, Olympus DAO itself has acknowledged that the kind of APY in its early days are unsustainable over the long term (see OIP-11), and has implemented a framework to calibrate the rebase rate downwards as the supply of OHM grows (see OIP-18).

Reducing the rebase rate may seem counter-intuitive, but there are benefits:

• Firstly, a lower rebase rate encourages bonding. The decision whether to bond liquidity and assets depends not only on the bond discount, but also whether it is more than the five-day rate of return from staking. So if the bond discount is less than that, it actually is better for the bonder to just buy and stake OHM directly. Hence, by reducing the rebase rate (which lowers the five-day rate of return), bonding can remain viable even with a lower bond discount. From Olympus’ perspective, this means that for the same amount of liquidity or assets that a bonder provides, the bond price can be higher. This translates into higher profits, which will eventually benefit stakers as the treasury grows.

• Secondly, a lower rebase rate can be seen as a form of “delayed gratification”, which will increase the runway at which Olympus DAO can sustain its rebasing rewards to stakers. This runway is a key metric that Olympus DAO tracks, and at the time of writing, it is able to sustain its current rebase rate for around a year, even if the protocol were to completely stop earning any revenue (the runway is updated in real-time on this chart). A longer runway through a lower rebase rate will provide more certainty to new investors, encourage them to buy and hold OHM for the long term, and enhance stability in the market for OHM.

In sum, the yield from staking OHM is not a scam, but the high APY figure is not meant to be permanent by design.

• Through bonding, Olympus DAO is able to profit from those want to buy OHM—even though they obtain their OHM at a slight discount to market price, Olympus DAO locks in a profit as well due to the bond premium. These profits from bonding provides the basis for Olympus DAO to reward stakers, through the new OHM that is distributed to them at every rebase.

• As Olympus DAO starts to mature as a protocol, it has rightly started to transit from trying to grow quickly, to now trying to grow sustainably. This entails increasing the runway for the protocol to sustain its rebasing rewards (i.e. yield from staking) via reducing the rebase rate. Doing so will also help maintain demand for bonding.

https://twitter.com/wagmianon/status/1473065750533390340

The power of staking

Even if one can be assured that OHM’s staking yield is sustainable, there is still the issue of what can happen if Olympus DAO’s flywheel starts to turn the other way—stakers start to unstake, sell their OHM en masse, and the price of OHM falls significantly.

While volatility in crypto cannot be completely mitigated, Olympus DAO has designed its staking mechanism to offer some countervailing effects should market sentiment turn extremely bearish.

The key to this is that the rebase rate is designed such that it is inversely proportional to the staking rate. In other words, when more people unstake OHM, and the proportion of staked OHM out of the total supply decreases, the rebase rate increases. So when more people unstake OHM, the remaining stakers will actually benefit more.

The example outlined by Olympus DAO in the event of a hypothetical bank run is illustrative, which is based on data as of 15 September 2021:

• In this scenario, we assume that an overwhelming number of stakers would unstake, and the proportion of staked OHM collapses from 92% to 3.3%. This leaves only 55,000 staked (out of the total circulating supply of around 1.67 million OHM).

• We also assume that the inflow to the treasury also stops completely, as no one is likely to want to bond during a bank run.

• Finally, we assume that those last standing stakers bought in at a price of USD 500 per OHM. Their initial investment in total would therefore be USD 27.5 million.

• At the point of this hypothetical bank run on 15 September 2021, the total OHM supply (i.e. OHM in circulation and in Olympus DAO’s treasury) was 2,082,553. In addition, what Olympus DAO calls the “risk-free value” (RFV) was USD 47,041,833. This RFV comprises the dollar value of the stablecoins in the treasury, as well as a marked-down value of the liquidity owned by Olympus DAO, and essentially indicates the amount of funds that Olympus DAO guarantees to use to back OHM.

• By subtracting the total OHM supply from the RFV, we get the number of new OHM that Olympus DAO can continue to mint (i.e. 44,959,280 OHM), given that each OHM has to be backed by 1 USD.

• If those last standing stakers remain staked, they will eventually be able to obtain this additional 44.9 million OHM tokens through the rebase rewards, which Olympus DAO estimates would take roughly a year to pay out in full.

• At this point, if the price of OHM is at its intrinsic value of 1 USD, those last standing stakers would still be sitting on a sizable profit. The value of their entire stake of OHM would now be worth around USD 45 million, compared to their initial cost basis of USD 27.5 million.

That said, this hypothetical bank run scenario is unlikely to play out. After a certain point, people will realise that extremely high rebasing rewards are being paid to stakers, and will therefore want to buy and stake OHM for themselves to capture this yield. Given this inherent dynamic, any bearish trend should not play out indefinitely.

Not surprisingly, the proportion of OHM staked has remained consistently high—at over 90%—since the launch of Olympus DAO.

It is simply in everyone’s interest to remain staked, which forms the basis of the “(3,3)” meme that Olympus DAO popularised across the crypto community.

(5) Is it worthwhile to buy OHM at such a high premium?

At this point, you might be convinced that Olympus DAO’s protocol is sound and want to buy some OHM yourself. But you might also be unsure about the extent of the premium you should be paying over OHM’s intrinsic value of 1 USD—given the potential rewards and risks, at what point would OHM’s premium be unjustified?

While the answer to this question will largely depend on your individual risk appetite and investment horizon, there are some guiding principles which can help you make a decision.

Expect both the rebase rate and price of OHM to trend down

Firstly, you should not be investing in OHM with the expectation that the rebase rate/APY will remain constant over time. As mentioned earlier, in order to prioritise sustainable growth, Olympus DAO has implemented a framework to calibrate the rebase rate downwards as the supply of OHM increases.

Secondly, as the rebase rate falls, it will only be natural for the premium for OHM to decrease. Hence, over the long term, you should expect that the price of OHM will trend downwards as well.

Consider the risk-free value (RFV)

Now, since the rebase rate and price are expected to keep decreasing over time, what benchmark can you use as a guide to determine the premium that you are comfortable paying for each OHM?

If you want to take the most conservative benchmark possible to determine the absolute minimum that your investment in OHM may be worth, you will have to consider the prospect of OHM’s price reaching its intrinsic value of 1 USD.

This is the basis of the risk-free value (RFV):

• If you recall, the RFV indicates the total amount of funds that Olympus DAO guarantees to use to back OHM, and it comprises the value of the stablecoins in the treasury, and a marked-down value of the liquidity.

• Another way to look at the RFV is that it should represent the total number of OHM that can ever exist if revenue stops completely, since Olympus DAO can keep minting new OHM until 1 OHM = 1 USD.

• By extension, OHM’s lowest possible market capitalisation (price multiplied by supply) would also be the RFV of Olympus DAO’s treasury.

So when you buy and stake OHM at any point in time, you have to consider that the theoretical minimum that your investment can go is actually the percentage of the RFV of Olympus DAO’s treasury that you capture then.

• After all, when you buy and stake OHM, what you are doing is to effectively capture a certain percentage of OHM’s market capitalisation.

• In the absolute worst-case scenario, even if OHM’s price reaches its intrinsic value of 1 USD, the total supply of OHM should eventually still reach its RFV, since Olympus DAO’s treasury has sufficient assets to back that number of OHM tokens.

• In this case, you should always maintain the share of the RFV as long as you continue to keep your OHM staked, and accrue your share of the newly-minted OHM until supply of OHM is equal to the RFV.

This means that for your investment in OHM to become truly “risk free”, the RFV of Olympus DAO’s treasury must eventually surpass the market capitalisation of OHM that you initially bought in. To illustrate:

• The current total supply of OHM is around 7.7 million, and the current price for each OHM is about 325 USD. This will put OHM’s current market capitalisation at around USD 2.5 billion.

• Let’s say I buy and stake 77 OHM now, paying 25,025 USD. This would allow me to capture 0.001% of OHM’s current market capitalisation, as well as RFV.

• For my investment to be truly risk free, I must be guaranteed at least 25,025 OHM, given the intrinsic value of each OHM to be 1 USD.

• The only way this can happen is when the total RFV of Olympus DAO’s treasury becomes more than USD 2.5 billion – the initial market capitalisation of OHM that I bought in at.

The multiple of OHM’s current market capitalisation against the RFV (let’s call it MC/RFV in short) is therefore a useful benchmark for you to assess whether OHM’s premium is justified. The lower the MC/RFV multiple is, the closer you are to making a truly risk-free investment.

For context, the MC/RFV multiple is sitting at approximately 12.5 now, if we take OHM’s current market capitalisation to be around USD 2.5 billion, and its RFV to be around USD 200 million. (Note that I have not been able to get fully accurate data, as Olympus DAO is in the midst of migrating to V2, resulting in its dashboards not being completely synced yet. This V2 migration is essentially an upgrade of its smart contracts to introduce new features.)

Consider non-RFV growth

At the same time, you should note that using the RFV as the basis for valuation is highly conservative, given that the market value of assets in Olympus DAO’s treasury is much more than the RFV. At about USD 700 billion currently, it is more than 3 times the RFV (likewise, data may not be fully accurate due to the V2 migration).

While this non-RFV portion of Olympus DAO’s treasury is not entirely risk free, this is where its growth potential may be the strongest moving forward. Olympus DAO has been embarking on various initiatives to diversify its revenue stream and drive further growth in its treasury:

• “B2B sales”: Olympus DAO launched the “Olympus Pro” platform in September, leveraging its bonding mechanism as a service for other protocols to acquire their own liquidity. In exchange, Olympus DAO gets to develop an additional source of revenue (it takes 3.3% of bonding revenue), while also diversifying the asset base in its treasury. Its partners will also obtain OHM as a treasury asset and/or to support its own liquidity. Thus far, the rollout of Olympus Pro has been successful—at the time of writing, it has at least 38 partner protocols on Ethereum, and a few more on other blockchains. If this can be sustained, Olympus Pro will certainly support Olympus DAO’s mission of building OHM to be the reserve currency for DeFi.

• “Strategic acquisitions”: Olympus DAO has also been accumulating massive amounts of strategic tokens as part of its treasury assets through bonding and DAO-to-DAO swaps. Not only do these tokens provide the potential for yield, they also confer Olympus DAO a say in the governance of other protocols. Given its intent to use this influence to strengthen partnerships with these protocols, this should likewise help drive the adoption of OHM across DeFi.

• “Spinning off subsidiaries”: The launch of [REDACTED] Cartel as a complementary subDAO also represents another mode of organisation for Olympus DAO, in which it can spin off branches to pursue more focused objectives without detracting from its primary mission of cementing OHM’s place as DeFi’ reserve currency. In the case of [REDACTED] Cartel, this should position Olympus DAO well in the ongoing “Curve wars”.

https://twitter.com/B49Jme/status/1476010917846401035

• “Expansion to new markets”: Finally, Olympus DAO is in the midst of its migration to V2, which has expanded the use of OHM into other blockchains besides Ethereum. Through gOHM (a “wrapped” form of sOHM that does away with a constantly-increasing token balance), those who are staking OHM can essentially deploy their sOHM on other blockchains and take advantage of opportunities in the wider DeFi space. Much like a multi-national corporation, the implementation of gOHM—together with its liquidity-bootstrapping programme called Proteus—may have the potential to make Olympus DAO a truly multi-chain protocol, with OHM the “global” reserve currency for DeFi.

Even though all these initiatives may not necessarily accelerate RFV growth, they are likely to grow Olympus’ overall treasury, and more importantly, make OHM a strategic asset within DeFi.

On this note, I think it is important to keep in mind Olympus DAO’s long-term mission of making OHM a true store of value, whose value is tied to fiat currencies as little as possible. Right now, a substantial portion of Olympus DAO’s treasury is comprised of stablecoins, which means that the value of OHM is still somewhat linked to fiat currencies. That said, if Olympus DAO wants OHM to be an independently-valued reserve currency for DeFi, untethered to the vagaries of governments, then it must build up a broader stable of decentralised assets in its treasury—all of which will not contribute to RFV.

For these reasons, if you believe in Olympus DAO’s “end game” and think that OHM is going to occupy a more strategic place over the longer term, you might be willing to tolerate a higher level of risk, i.e. buy into OHM at a higher MC/RFV.

“Diamond hands” have value too

In addition, I do believe that conviction can also allow the more committed stakers to capture value too, given the way Olympus DAO has designed its staking mechanism.

Recall that the rebase rate is designed such that it is inversely proportional to the staking rate. So if more people unstake (which they have to do in order to sell their OHM), those who remain staked can enjoy a higher return.

To me, this makes it seem unlikely that the price of OHM will fall to its RFV in the short term, since the increased staking rewards should help cushion some of the price declines. Of course, this is entirely speculative, and it remains to be seen how the market for OHM will pan out when we enter a severe bear market for crypto.

Nevertheless, I do have a soft spot for investment theses that reward those with “diamond hands”. With OHM, even if sentiments turn bearish, I know that I can capture some value from those who capitulate as long as I remain staked. For this reason, I feel personally comfortable holding and staking my small stash of OHM for the long term, and the short-term price action of OHM will definitely not affect my sleep.

Understand the key risks

Of course, there are still significant risks associated with Olympus DAO and DeFi more generally. These include:

• Competition. Olympus DAO’s success has led many others to copy its design to create their own protocols, while offering higher yields. This fragments the available pool of assets and liquidity that Olympus DAO can acquire, and will curb its growth. Already, the size of Wonderland’s treasury has overtaken Olympus DAO’s, even though it began only in September. It remains to be seen whether Olympus DAO can sustain its value proposition amidst this competition, and whether there is enough space for multiple winners with different niches within DeFi (what the elusive Concave protocol is trying to achieve may be helpful in this regard).

• Smart contract risk. All DeFi protocols are built on smart contracts on blockchains. Given the nascence of the space, there may be loopholes in some of these smart contracts, leading to exploits or hacks. While Olympus DAO’s smart contracts are not immune to this risk, the fact that they have been audited by third-parties should lend some assurance. That said, given that Olympus DAO puts some of its treasury assets into other DeFi protocols to generate yield, it is also exposed to the smart contract risk of those protocols. As one of Olympus DAO’s contributors explained in this article, this risk scales not linearly but exponentially as Olympus DAO becomes more integrated with other protocols. Nevertheless, by diversifying its treasury yield strategies (and as the Defi space matures), Olympus DAO should be able to keep this risk manageable.

• Stablecoins losing their peg. The risk-free value (RFV) of Olympus DAO’s treasury is dependent on the stablecoins in there (i.e. DAI, FRAX, LUSD, UST) retaining their peg. As DAI and FRAX are at least partially collateralised by USDC and other assets, they may be affected if centralised stablecoins get impacted by hostile regulatory action. Meanwhile, even though LUSD and UST are decentralised, their algorithms to maintain the peg may not be immune to failure as well.

• Governance risk. Olympus DAO’s treasury resides in a multi-signature vault, and any transactions using the treasury’s assets must be approved by at least four out of seven total signers. While it does not seem likely that the core team behind Olympus DAO is out to do a “rug pull”, there is still a non-zero risk of insider collusion at the expense of the broader community. At the same time, while Olympus DAO is working towards more decentralised governance, this may present other forms of risks, e.g. a third-party with sufficient resources or influence could seize control of governance to benefit their self-interest.

• Black swan event. Finally, even though Olympus DAO’s protocol has structured its incentives such that staking is the most optimal strategy for all, a black swan event—such as a complete meltdown in the crypto markets—could deliver a terminal blow that the protocol may not be able to recover from.

Given these risks, investing in OHM will not be everyone’s cup of tea. This is only normal, for we all have our own personal considerations when it comes to our investments.

On the flip side, even if you really like what Olympus DAO is trying to do and a higher risk appetite, you should also not be going all in on OHM. Even if the potential rewards seem attractive, always be sure to diversify your investments. After all, nothing is ever certain in this world.

(6) Thoughts on Olympus DAO’s future

If you have made it this far, I think you can see for yourself how complex Olympus DAO’s protocol is. It took me a while to wrap my head around how Olympus’ bonding and staking mechanisms are supposed to work, and it was only until I took the time to write this essay that I can say with some confidence that I finally got how everything came together.

Perhaps, the simplest analogy I can offer to explain Olympus DAO is that it is a community-owned central bank—or a “Grassroots Federal Reserve” in the words of this article—with the following key traits:

• Through a virtuous cycle of bonding and staking, Olympus DAO draws in liquidity and assets into its treasury, and distributes the gains to its community in the form of OHM. This process can be sustainable if the parameters for bonding and staking are calibrated well—the potential reward from staking can create a latent demand for bonds, which will in turn provide the basis for the staking yield.

• Because Olympus DAO’s treasury is like a “black hole” in which its liquidity and assets cannot leave (unless OHM falls below its intrinsic value), OHM will always be backed by some value. As Zeus puts it in this Tweet thread, “A protocol that owns its own assets can go dormant but can never die.” Likewise, a currency that is backed by such a protocol should always retain some of its value.

• Governance of OHM will also reside in the holders of the actual currency, at least when Olympus DAO establishes a more decentralised form of governance after its migration to V2. Community governance may perhaps be the most experimental aspect of Olympus DAO’s grand experiment in creating a new monetary system within DeFi, and it will certainly be fascinating to see where this goes over the long term.

Ultimately, I lean towards the more hopeful camp with regard to Olympus DAO. The core team has executed methodically and finely, while growing a passionate yet measured community alongside this journey. Given this track record, I think Olympus DAO has what it takes to become a leading DeFi protocol, at least in the medium term. Furthermore, while many risks still abound, DeFi is still in its early days, and I can’t help but feel incredibly optimistic about Olympus DAO’s potential within this broader space.

Notwithstanding my personal investments into OHM, I am also rooting for Olympus DAO because I think it upholds the original ethos of DeFi—of democratising finance, and sharing value with a broader group of stakeholders. A “grassroots Federal Reserve” does sound like a lofty ideal now, but if Olympus DAO succeeds in pulling DeFi together through OHM, this will definitely be one for the history books.

Acknowledgements: I will like to thank Ewen and Liki for their insightful comments, which have allowed me to sharpen the assessment of the risks and growth prospects of Olympus DAO in this essay.

Disclaimer: Nothing in this essay constitutes financial advice. Please do your own research and be mindful of your own investment objectives before investing into any cryptocurrencies, DeFi protocols or NFTs.

Cover image by Spencer Davis from Pexels.

In the following sections, I will try to develop this analogy in more detail, as I think this will be helpful for those new to the crypto space—to acquire a top-level overview of its key “attractions”, and thus feel more confident to step in and wander.

Blockchains as “countries”

Before we start, we should be clear what blockchains are. I have tried to outline their defining properties in my previous essay. But to summarise, blockchains are essentially databases on a distributed network, which can allow anyone to add entries to (permissionless), and whose data can be seen by everyone (transparent) but cannot be changed once it is validated by all nodes (immutable).

On top of that, what makes blockchains akin to countries are two additional factors:

• Most blockchains have their own native cryptocurrency, which is needed to pay for any transaction or activity on the network—just like how you would need Euros if you want to do things within the European Union (with the exception of the UK).

• Many blockchains also allow you to programme smart contracts, which are basically code that takes in a defined input (e.g. a specific amount of cryptocurrency) and delivers a defined output (e.g. minting a specific non-fungible token or NFT). Such smart contracts allow for more complex applications to be run on a blockchain (beyond simply sending cryptocurrencies to one another), and these applications can be somewhat analogous to the specific attractions you visit and engage with when you travel to a new country.

The major blockchains with smart contract functionality and an established ecosystem of applications that you can now readily enter and explore include Ethereum, Polygon, Solana, Terra, and Avalanche. (This list is non-exhaustive, but I have only outlined those that I am familiar with.)

Before you leave

Like travelling, getting into the crypto space requires some prior preparation. At the very least, you need to ensure that you have the right documents to travel (i.e. your passport) as well as some cash in the relevant currency.

Getting your documents in order

The equivalent of a passport in the crypto space is known as a wallet, which usually takes the form of a browser extension or a mobile app. Your crypto wallet is the interface by which you verify your ownership of crypto assets (e.g. cryptocurrencies and NFTs) on a blockchain, and interact with applications built on it.

However, unlike a physical wallet, the crypto assets you own are not actually in your crypto wallet, but remain within the blockchain. Your crypto assets are simply tied to the public address of your crypto wallet, and through public key cryptography, you can prove that you control the wallet address and hence those assets.

It is useful to understand how this works, so that you can use your crypto wallets with care when navigating the crypto space, just as how you would take care of your own passport:

• When you set up a crypto wallet, a private key—a string of letters and numbers—is randomly generated, which may be presented to you in the form of a seed phrase. This is the only thing that enables a user to control all the assets tied to the particular wallet, i.e. cryptocurrencies and NFTs. This is why your private key or seed phrase must be kept in a secure location, somewhere not connected to the internet (e.g. on a piece of paper or dedicated hardware wallets like those from Ledger). Losing one’s private key or seed phrase is arguably even more catastrophic than losing one’s passport, as it cannot be replaced or recovered by anyone, including the wallet service provider.

• The private key, in turn, is used to derive a public key through a one-way function called the Elliptic Curve Digital Signature Algorithm or ECDSA. Without getting into the mathematics behind it, the outcome is that while the public key can be derived from the private key, it is next to impossible to derive the private key from the public key. Much like an email address, your public key is what applications on the blockchain “see” when you interact with them via your wallet.

• Your private key and public key work in tandem when you make any transaction on the blockchain. Your wallet will allow you to sign the transaction data with your private key, which creates a signature using ECDSA that essentially proves you own the private key without revealing it to anyone.

Another important thing to note about crypto wallets is that most of them are currently not interoperable across all blockchains.

• The most established wallet now is MetaMask, but it can only be used to interact with the Ethereum blockchain, Ethereum-compatible blockchains like Polygon and Avalanche, as well as Ethereum roll-ups such as Arbitrum and zkSync. (Rainbow is another Ethereum wallet worth checking out, with a more vibrant user interface than MetaMask.)

• For Solana, the leading wallet seems to be Phantom, while for Terra, you have to use the official wallet called Terra Station.

• Nevertheless, cross-chain wallets that will allow you to operate across most of the established blockchains are being developed. An example is XDEFI.

• In addition, there are also crypto wallets that aim to make the user experience as simple as possible. For instance, Torus allows you to easily set up a wallet using your social media accounts (e.g. Google and Facebook), and it manages your private key in the backend through distributed key generation.

Getting the currency you need

Once you have the required crypto wallet(s), you will need to convert your fiat currency into the relevant cryptocurrencies. To do so, you will have to use what is called a fiat on-ramp, which is akin to a money changer to convert your local currency into the foreign currency you need.

The most established fiat on-ramps are your crypto exchanges, such as Coinbase, FTX, Crypto.com, and Gemini. Which exchange is best suited for you will depend on a few factors:

• Where you live. Each country has their own set of regulations governing crypto-related activities, which may result in certain exchanges being unavailable to users from particular countries. For example, Binance.com was recently ordered to stop operations in Singapore.

• What cryptocurrencies you need. The cryptocurrency you need may not be available on every exchange, so this is something you have to double-check on the respective exchanges’ website. Nevertheless, the established exchanges should offer most of the key cryptocurrencies you will need, e.g. BTC, ETH and USDC.

• What blockchains you want to use. This is an often overlooked aspect of selecting which crypto exchanges to use. Some exchanges may allow you to trade certain cryptocurrencies, but do not allow you to transfer them out to an external wallet. Others may allow you to transfer them out, but only to limited blockchains. For example, while I can buy MATIC or LUNA on Gemini, I currently can only transfer them to a wallet on the Ethereum blockchain, and not to their native blockchains, i.e. Polygon and Terra respectively. (If I need MATIC or LUNA on their native blockchains, I use Crypto.com instead.)

• Transaction fees. Crypto exchanges do not operate for free, and will levy various transaction costs, e.g. to trade and withdraw cryptocurrencies. This is therefore something you should check before you sign up to use an exchange. Do also note that some exchanges such as Coinbase and Gemini have interfaces for more advanced users, which offer much lower trading fees as compared to the default interface. For instance, trading fees on Gemini’s ActiveTrader interface are currently around 0.25%-0.35%, while those using its default web or mobile interface will have to face trading fees of around 1.49%. Since the savings can be rather significant, I highly suggest going straight to using these more “advanced” interfaces instead; they are not that difficult to use, and the experience will be very similar to how you buy a stock (e.g. setting limit orders).

During your trip

Once you have your crypto wallet(s) set up and transferred over some cryptocurrencies into them, you are now ready to enter and start exploring the crypto space properly.

Guides

Given the complexity of blockchains, it makes sense to consult guides to help you navigate the space, much like how you may rely on tour guides and/or guidebooks such as Lonely Planet when travelling.

If you are a complete beginner, I recommend investing time in an online course that covers all the basics. I did a couple of courses on Coursera about blockchain business models and decentralised finance, which gave me the confidence to dive deeper into the crypto space. These courses are completely free to audit, but you have to pay if you want a certificate. (Since the crypto space evolves very quickly, it is possible that the materials in a certain course may be outdated by the time you view it. Hence, do check on when a course was created or updated before jumping into one.)

If you already have some background with regard to blockchains, the important thing is to keep up with the rapid developments in this space.

• Good places to start include crypto-focused publications such as CoinDesk, Cointelegraph and Bankless.

• The venture capital firm a16z has also compiled extremely comprehensive sets of readings on the crypto space in general, as well as on non-fungible tokens, and decentralised autonomous organisations specifically. They are all worth bookmarking and perusing.

• For visual learners, YouTube has channels like Whiteboard Crypto and Finematics that create bite-sized video explainers on crypto-related topics.

• One crypto education project that has piqued my interest is Curious Addys’ Trading Club, which aims to create a “Duolingo for crypto”. Through a game (which is still being developed), users can learn how to trade cryptocurrencies and NFTs in a risk-free environment with a supportive community.

• For those who enjoy numbers and lists, Defi Llama aggregates and ranks applications across most blockchains based on the total value of digital assets locked within them. Their rankings provide a good overview of the top blockchain-based applications being used right now.

Finally, if you want to follow developments in real time, there is no other way but to follow the projects you are interested in (and their core team) on Twitter and Discord. Most developments are first highlighted on these two platforms, and being on them is a must if you want to go deep in your crypto “travels”.

Routine expenses

Just as how you will have to set aside some cash for expenses when travelling (e.g. buying food or taking the bus), you will have to do likewise when exploring the crypto space due to something called gas fees.

These fees are essentially transaction costs for using the blockchain, and you can think of them as payment to the blockchain’s node operators for devoting computational resources to validate your transactions.

Due to high demand for applications on the Ethereum blockchain, gas fees there have been rather costly. I have personally experienced gas fees ranging from the equivalent in ETH of US$10 to US$150, depending on the complexity of the transaction. (This article provides a concise explanation of what gas fees entail, and how to minimise them when using the Ethereum blockchain.)

Fortunately, gas fees on most other blockchains are significantly lower. For instance, gas fees on Polygon are almost always comfortably below 0.01 MATIC (~US$0.02). Hence, many applications on Ethereum have migrated to the so-called “Layer 2” blockchains, or are in the process of doing so.

Souvenirs

While travelling, I’m sure most of you will like to buy some souvenirs to showcase or share with your family and friends. You can also do something similar in the crypto space with non-fungible tokens (NFTs).

As the term “non-fungible” indicates, NFTs are essentially tokens on a blockchain that are unique, i.e. no NFTs will be exactly the same. This is in contrast to cryptocurrencies, which are fungible, i.e. one BTC or ETH is always equal to another BTC or ETH.

The primary use of NFTs thus far has been to represent ownership over unique digital assets such as various forms of art and media. Since records on a blockchain cannot be tampered with, NFTs provide a permanent form of assurance that their holders are the legitimate owners of the underlying assets. Even if those underlying assets can be easily duplicated (since they are digital), the NFTs themselves cannot. Hence, there can only always be one or a fixed number of owners over those assets—much like how even though there are many replicas of the Mona Lisa painting, there is only one legal owner of the original painting, i.e. France. (Note that the legal status of NFTs is still ambiguous, and the mainstream legal opinion is that NFTs do not grant the owners copyright over the underlying asset.)

In any case, NFTs have heralded an explosion in transactions of digital art and media. Creators of these assets can now sell them directly to collectors via NFTs, reducing the influence of middlemen and allowing for closer relationships with their fans. As some have put it, NFTs allow for “patronage+”, in which supporting creators becomes no longer just an act of altruism, but also an investment. (For a really deep dive on the value of NFTs, check out this tour de force of an essay.)

My personal view is that it wouldn’t hurt to purchase an NFT or two if you find something that resonates with you as you explore the crypto space. Even if the NFTs no longer commands the same monetary value you paid for it, it can serve as a souvenir of your time spent exploring the crypto space at the very least—representing a creator or project that you are proud to support and showcase to your friends.

In particular, I am interested in NFTs that can help support our fight against climate change, and I recently bought some NFT “rocks” minted with tokenised carbon offsets (see first Tweet below). I hope this can mark the beginning of a personal learning journey into the intersection of carbon markets, crypto and art. (See also the second Tweet below for an NFT that actually locks in more carbon offsets over time, but I couldn’t afford the auction.)

https://twitter.com/CarbonizedNFT/status/1462894392272777221

https://twitter.com/s_v_n/status/1466846151903121408

Separately, I also bought a couple other NFTs simply on the basis of their aesthetic quality (which of course is subjective):

• I minted witch #7441 from Crypto Coven’s public sale that ended recently. I was drawn not just to the intricate and textured artwork, but also the finely-crafted narrative surrounding the entire project. The tone was absolutely on point, and the words they used melded together beautifully (go read through the AI-generated “articulations” of each of the witches on the website). It is clear from the execution of the entire project that the team has good taste—even its smart contract code has been praised as a “work of art”!

• I bought Genesis Deity #411 from Neon Pantheon. The premise of having deities that amalgamate features from different cultures and mythologies is intriguing, and the project’s artwork looks pretty slick overall as well. Even though minting is still ongoing, I decided to buy a Genesis Deity off OpenSea because I wanted one whose features I can identify with (rather than take the chance with the mint).

• I was also fortunate to mint a “rare” piece from JPG People Club (but not so fortunate to mint the “super rare” or “OG” ones). The concept is fun—you get to input some text of your choice when minting the pieces—and the artwork is pretty cute as well.

While I certainly don’t expect to profit off these NFTs, I see them more as a way to put some skin into the game, and force myself to stay connected to this space. More importantly, I want to keep an open mind for even more innovative uses of NFTs that are bound to arise. There is an incredible amount of creative energy devoted to the space (as with a lot of spam and scams), and we are certainly in for many more surprises in the future.

Ticketed attractions

Besides being souvenirs, NFTs can also function somewhat like tickets to attractions that you want to visit.

Certain blockchain-based applications may require ownership of specific NFTs for you to engage with them. This is most evident in the fast-growing space of blockchain-based games, in which you can own in-game characters and assets as NFTs.

For example, Axie Infinity, which is one of the most popular blockchain-based game at the moment, requires players to own at least three “Axie” NFTs to play the game. Each Axie is a creature (somewhat like a Pokemon) with a unique combination of attributes that can influence gameplay. Players lead their teams of Axies to complete missions against computer-controlled enemies or battle other players to earn the in-game cryptocurrency. This can be used to “breed” new Axies or be exchanged for other cryptocurrencies, which is why Axie Infinity has also been characterised as a “play-to-earn game”.

Despite the potential to earn some money, there are still limitations to such blockchain-based games. For one, the cost of purchasing the required NFTs just to play the game may be prohibitive. Arising from this, some have criticised games like Axie Infinity as a “pay-to-win” game, given that Axies with more powerful attributes command far higher prices. In addition, many of these blockchain-based games are still rather primitive and not exactly fun, as compared to “normal” games already in the market. Nevertheless, this is still a nascent space, and I believe that the ability to own and control your own in-game avatars and assets (perhaps even across different games) will be a powerful driving force for blockchain-based games as a whole.

Besides Axie Infinity, other blockchain-based gaming projects that have piqued my interest at least somewhat cursorily (I don’t consider myself an avid gamer) include:

• The Sandbox allows people to buy digital land (as NFTs), and build their own virtual experiences, including games, on them using assets (again as NFTs). Prices of the Sandbox’s native token (SAND), alongside other similar projects, have surged after Facebook’s rebranding as Meta spurred interest in the ”metaverse”.

• Gala Games has a range of blockchain-based games under development, which reprise popular genres such as tower defense, simulation and fantasy RPG.

• Blankos Block Party has one of the more polished and developed gameplay of the blockchain-based games currently out there, while also being probably the least crypto-centric. Besides the player-controlled avatars called “Blankos” which are NFTs, there is no other token-based economics that are common in other blockchain-based games. Players simply control their Blankos to partake in multiplayer mini-games with different modes such as racing and shooting. The game studio behind this game, Mythical Games, has been recently valued at more than US$1 billion.

• Crypto Raiders is a game with classic sprite-based 2D graphics (like those in the Nintendo Game Boy Color I grew up playing) in which players can send their characters to raid dungeons and earn loot. The distinctive element in this game is that your NFT characters can “die”, resulting in them becoming unplayable within the game. This adds an element of risk that is perhaps unmatched by other games.

• Smol Brains is another game with sprite-based 2D graphics. Players have to stake their monkey characters (which are NFTs) to earn IQ points, and what is noteworthy is that the images of those monkey characters will dynamically change as they grow in intelligence (i.e. the size of their heads increase). This will unlock new tech breakthroughs for the players’ characters, and enable them to do more things within the “Smolverse”.

Investing

If you come across a compelling investment opportunity while travelling (e.g. a property with an attractive selling point or an interesting business), it will probably be quite difficult to directly invest in it as a foreigner. This may be due to local laws and the lack of trustworthy local intermediaries.

Such concerns would be less of an issue in the crypto space, since the permissionless nature of blockchains means that you can easily and securely move your capital around to take advantage of investment opportunities across different applications and even blockchains. This is what decentralised finance (DeFi) is about.

The central value proposition of DeFi is to reduce the need for traditional intermediaries like banks for certain financial services. Instead of these intermediaries, DeFi applications powered by self-executing smart contracts on a blockchain enable users to engage in services like lending/borrowing, trading and asset management, often at a lower cost and with lower barriers to entry.

Hence, even without using leverage, annualised yields on stablecoins (cryptocurrencies pegged to fiat currencies) can easily be at least 5%, which trump the interest rates offered by traditional banks.

In fact, the Anchor protocol on the Terra blockchain currently offers a mindboggling 19% APR (annual percentage yield) for simply depositing the TerraUSD stablecoin (UST) into its application. This yield is in part obtained from both the interest paid by those borrowing UST from the Anchor protocol, as well as the staking rewards generated by the assets placed by these borrowers as collateral (which are currently ETH and LUNA). [Afternote: UST and LUNA collapsed in early May, suffering from a classic bank run. Rather than an indictment of DeFi as a whole, this had more to do with how UST was designed, i.e. it was stabilised by an endogenous asset which led to a higher risk of a negative feedback loop when confidence in either UST or LUNA fell. Notwithstanding the resultant impact on crypto markets, I still believe in DeFi’s central value proposition of reducing intermediary costs and barriers of entry for financial services.]

https://twitter.com/GabrielGFoo/status/1469331215295533059

Of course, there are still significant risks associated by DeFi applications, such as the risk of a smart contract being exploited or governments outright banning them. But I think it will be very hard to stop the sheer economic logic of lower transaction costs and higher yields. Furthermore, these risks will be mitigated as the space matures, through improved smart contract audits and DeFi insurance, as well as—eventually—greater regulatory clarity from authorities.

The crypto asset management firm Bitwise anticipates that the DeFi industry is poised to grow to several times of its current market capitalisation (around US$150 billion), especially when compared against the US$24 trillion market value of the traditional finance sector.

I agree that DeFi is likely to take up a non-trivial share of the financial industry as a whole, and I would not be surprised to see traditional banks or fintech companies offering services based on DeFi in the years ahead—somewhat like how Coinbase recently gave its users with the DAI stablecoin the option to earn yield through the Compound protocol.

Finding community

Of course, the primary motivation behind why we travel is less often about material things (e.g. buying souvenirs or making investments). It is the intangibles of travel—the relationships you make and the communities you experience—that form our most treasured travel memories, the ones we cherish most deeply.

Likewise, the crypto space is so much more than just trading the hottest cryptocurrencies and NFTs, or seeking out the highest-yielding DeFi projects. Fundamentally, it is about people. You come for the gains, but stay for the community.

From my explorations in the space thus far, I have been repeatedly amazed by the people who have conceived and built many fascinating applications using smart contracts and blockchains, as well as how thriving communities have emerged and coalesced around these use cases. Many of these communities take the form of decentralised autonomous organisations (DAOs).

While many definitions of DAOs have been put forth, I don’t think that there is a firm definition yet given their nascence. Broadly-speaking, we can conceive of DAOs as blockchain-enabled online collectives, in which the permissionless and transparent nature of blockchains are harnessed to empower groups of people to work together towards shared objectives.

https://future.a16z.com/dao-canon/

To me, the key differentiating factor between DAOs and other forms of human organisations is that certain processes within a DAO can be administered via smart contracts on a blockchain, without members necessarily having to completely trust each other. This provides DAOs with some unique characteristics:

• Operations. The smart contracts underpinning relevant aspects of a DAO’s operations (e.g. how the DAO uses crypto assets deposited into its treasury) are self-executing and transparent to all. Members can therefore be assured that any pre-determined rules are applied automatically and consistently.

• Governance. Core decisions regarding the DAO can also be subjected to the consensus of all members. With blockchains, voting can be done on-chain in a secure, permissionless and transparent manner, with voting rights restricted to only holders of the DAO’s governance token, which can be earned via contributing to the DAO. Innovations like quadratic voting can also help mitigate the drawbacks of existing voting mechanisms, i.e. “one dollar, one vote” (which over-privileges those with money) and “one person, one vote” (which under-privileges those who are able to contribute more).

• Ownership. Perhaps most importantly, DAOs can democratise ownership. Unlike traditional companies in which equity is concentrated in the hands of founders and investors, DAOs can enable a much broader range of stakeholders to have a share, such as users, contributors, vendors and other community partners. By spreading out ownership among the very people who create value in them (e.g. through tokens), DAOs may be in a better position to bring about the virtuous cycle of stakeholders being rewarded for their contributions, and in turn, becoming even more motivated to work hard for their DAOs’ interests.

As I mentioned in my previous essay, the range of DAOs that have sprung up is truly astounding, and examples include DAOs that aim to become a venture capital fund (e.g. the LAO), administer a crypto-collateralised stablecoin (e.g. MakerDAO), create a decentralised reserve currency (e.g. Olympus DAO), collect digital art (e.g. PleasrDAO), create an exclusive cultural community (e.g. Friends with Benefits), facilitate philanthropic grants (e.g. Big Green DAO) and even drive up the price of carbon offsets (e.g. Klima DAO).

Granted, we’re still in the early days of DAOs, and much remains to be seen whether they can scale up and find success. DAOs will have to navigate the same risks and pitfalls that afflict any human interactions, i.e. ego, greed and what not. Nevertheless, the potential DAOs bring in terms of facilitating trustless cooperation is unprecedented, and I am personally excited to see the things that they can accomplish in the years ahead.

Travelling onwards to another country

Finally, most of you will probably not be content with simply visiting one country when travelling—you want to travel to as many countries as possible. This is not an issue in the crypto space, given the permissionless nature of the established blockchains.

In addition, unlike the sometimes excessive baggage fees charged by airlines, the cost (and effort) of moving your digital assets across different blockchains tends to be more manageable. The way to do this is through something called a bridge.

How such bridges work is generally through a “lock and mint” and “burn and release” mechanism:

• Let’s say I want to transfer 1 USDC from the Polygon blockchain to the Avalanche blockchain. What a bridge does is first to lock 1 USDC in my wallet on the Polygon blockchain through its smart contract,. Then, it will mint (i.e. create) 1 USDC on the Avalanche blockchain, which will be credited to my wallet there.

• Should I want to transfer 1 USDC back from Avalanche to Polygon, the bridge will burn (i.e. destroy) 1 USDC from my wallet on Avalanche, and release the original 1 USDC to my wallet on Polygon.

• Through this process, the total number of the asset being transferred remains the same across both blockchains, and no one can use the asset on both chains at the same time.

Some of the more established bridges include AnySwap, AllBridge, Wormhole, and Synapse Protocol. Note that not every asset or blockchain may be supported by these bridges, but more options will certainly be offered with time. In addition, while bridges are mostly used to move cryptocurrencies across blockchains now, some of them support the movement of NFTs as well, e.g. Wormhole.

Conclusion

In this “travel guide”, I have tried to provide a broad overview of all the key elements that one would encounter when navigating the world of blockchains—from essential infrastructure like crypto wallets and fiat on-ramps, to key attractions like NFTs, DeFi and DAOs.

It turned out to be much longer than I had initially anticipated, which reflects the sheer breadth and complexity of the crypto space. Even then, I feel that I have barely scratched the surface in terms of both the underlying mechanics of blockchains, and the immense possibilities of their applications.

Needless to say, given how nascent the entire space is at the moment, it is very likely that some of the things I have written above will become obsolete in the months and years ahead. In fact, I foresee that in the not-too-distant future, one’s interactions with blockchain-based applications will be so seamless that you need not even know that you are dealing with blockchains in the first place (see for example Shopify’s recent launch of its NFT Beta Program that allows businesses to mint and sell NFTs directly on its platform).

That said, I believe that we should all try our best to understand the basics of blockchain technology today, so that we can be in a good position to appreciate and take advantage of its revolutionary possibilities in the future. Whether it is in the realm of investing or entertainment, or simply the ever-present human impulse to seek community, my hunch is that blockchains will redefine many of these aspects of our lives.

Hence, as we peer into this emerging crypto space, I hope we can all also hold on to the same spirit of curiosity and openness that often undergird our most transformative travel experiences in the real world. The path ahead will be fraught with much risk and volatility, but I think the experience itself will be worth it. Even if I do not get enriched financially or materially from this journey, I am confident that I will gain a much richer understanding and appreciation of our constantly-evolving world, and the people who make it what it is.

Bon voyage!

Disclaimer: Nothing in this essay constitutes financial advice. Please do your own research and be mindful of your own investment objectives before investing into any cryptocurrencies, DeFi protocols or NFTs.

Cover image by Valentin Antonucci from Pexels.

I first heard of blockchains and cryptocurrencies in 2017 during the previous Bitcoin bull run. While I made sure to get a cursory understanding of how blockchains worked before buying a small amount of cryptocurrencies then, I had treated these purchases as pure speculation. I couldn’t imagine using bitcoins as part of my daily life, and Ethereum’s smart contract functionality seemed like a rather lofty aspiration at that point in time. It was just fun to join in the hype (with a relatively inconsequential stake), and see where everything went.

In the years after, I did not pay any regular attention to the space. I simply chucked the little bits of cryptocurrencies I had into a mobile wallet app and left them alone. I did not bother myself with tracking their prices, nor harbour much interest to learn about what developers were building on blockchains more generally. Bitcoin was very much like loose change left behind in my pants pocket.

Over the past year, however, their “jingling” became louder. Of course, the massive price appreciation of many cryptocurrencies since end-2020 was a factor. But what really sparked my intellectual curiosity about the crypto space was the explosion in use cases today and the sheer creativity underpinning them. This fact was not evident to me at first, until bits and pieces gradually started percolating through my non-crypto information diet this year—in large part thanks to Packy McCormick’s Not Boring newsletter on business strategy.

Eventually, I had the niggling sense that blockchains, cryptocurrencies and what is called “web 3.0” are now phenomena that cannot and should not be ignored. So about a month ago, I decided to try to learn about the crypto space more intentionally, and embarked on a couple of online courses on Coursera about blockchain business models and decentralised finance (or DeFi in short). In the process, I also started to play around with some DeFi applications, follow “Crypto Twitter”, and lurk in the Discord servers for a couple of decentralised autonomous organisations (or DAOs).

I still think this is a space rife with many unknowns and risks. But I also emerged from these early explorations with a newfound fascination and optimism about the possibilities it can offer. I therefore wanted to consolidate my learnings and thoughts thus far in this essay, on a platform that allows one to publish on a decentralised database no less, and convince myself (and perhaps you the reader) that crypto is now worth one’s investment—not necessarily in money per se, but definitely in time and attention.

The defining properties of blockchains

To start, it is imperative to have a clear understanding of what blockchains actually bring to the table, given the dizzying variety of applications being built on blockchains today, with new projects almost popping up like weeds all the time.

Essentially, blockchains are databases—in fact, you can think of a blockchain as something like an Excel spreadsheet containing information like transaction data or code to run basic programmes—but with some additional defining properties.

The first of these properties is that blockchains are distributed across a network. Each node in the network holds a copy of the database, and the collection of all these copies are what makes up the blockchain. This differs from an Excel spreadsheet hosted on the cloud, which while accessible to many, is still ultimately based on a single “master” copy of the document residing in a central server or two (for back-up). In contrast, the copies of the database in every node are in a sense equal, and so if any node goes down or becomes corrupted, the entire blockchain network itself will not be impacted. This is how blockchains can be highly resilient and decentralised; Ethereum for example is made up of thousands of nodes.

This brings us to the second key property of blockchains in that they can be completely permissionless. Given that a blockchain is distributed with no “master” copy of its database, any addition to the database will require a consensus among the nodes in the network. The exact threshold and process of achieving consensus differ by design, but the most significant implication of these consensus mechanisms in my opinion is that they allow for a network to be decentralised while remaining open, i.e. one can set up a node and start validating transactions on the blockchain without seeking permission from any gatekeeper. All you need to do is to meet the basic hardware requirements and undertake to participate in the network’s consensus mechanism. (Note that there can also be private or permissioned blockchains in which only designated parties can participate, but I suspect that the key selling point for blockchains compared to existing platforms is the capability of creating permissionless networks.)

The next key property of blockchains is their immutability. Once a set of transactions are validated through the blockchain’s consensus mechanism (as a block) and then “chained” to the database (hence blockchain), they cannot be edited. A malicious node cannot simply amend the database due to some nifty cryptographic features. In general, how this works is that at the end of each block of data within a blockchain, there is a hash that is derived from a one-way mathematical function of all the data within the block. This hash is then repeated at the start of the next block of data, and effectively informs how the different blocks of data have to be ordered within the entire blockchain. The important point is that if even a single bit of data changes, the hash will become completely different, and the blockchain becomes “broken”. In such a case, the network will simply replace the corrupted block with the original copy of the blockchain from other nodes.

For blockchain networks to be able to achieve consensus and remain immutable, they also have to be transparent—the final key property of blockchains. This means that every node, or effectively anyone for permissionless blockchains, must be able to see all the data that have been ever locked into the blockchain, so that they can validate new transactions and prevent any changes to old ones. Indeed, the popular blockchain mantra, “don’t trust, verify”, is only possible if the entire network is transparent.

Taken together, I believe this distributed, permissionless, immutable and transparent nature of blockchains is what sets them apart and accounts for their revolutionary promise. By leveraging on these properties, decentralised applications (or dApps) that are being built on top of blockchains can create value for users in ways that may not be easily replicated by non-blockchain systems.

Reducing the need for intermediaries

The most obvious value proposition for blockchains is that they can reduce the need for intermediaries in the provision of certain services. As a permissionless network, blockchains allow users to interact with each other in a secure manner without the need for any third-parties to process the interaction. Instead, the rules governing such interactions can be set out in a smart contract embedded within the blockchain, which is transparent to all users and cannot be changed by anyone. By effectively removing the middleman, dApps utilising such smart contracts can thus reduce transaction costs for users.

The world of DeFi offers perhaps the clearest manifestation of this. By now, I think most of us will be aware that we can easily send cryptocurrencies over a blockchain network to another person without a bank account. But DeFi takes a step further beyond payments, and allows one to undertake other financial services without a bank as well.

Take lending and borrowing for instance, a core function of traditional banks. DeFi protocols such as Compound and Aave now allow you to do this without any intermediary. For lenders, you can easily deposit crypto assets into their applications and immediately start earning interest on them. For borrowers, you can likewise deposit your collateral (as crypto assets) and take out a loan against them. Each crypto asset will have a certain loan-to-value limit or collateral factor that determines the amount of the loan a user can take out. As an illustration, Aave’s loan-to-value limit for the USDC stablecoin is currently 80%, so if I deposit 1000 USDC into the application, I can borrow crypto assets worth up to 800 USDC.

Where this gets interesting is that if at any point my loan becomes under-collateralised (due to price changes in the crypto assets underlying my collateral or loan), the smart contract governing the application will allow for my loan to be liquidated by other users. For a lender, this effectively eliminates the risk of a default from the borrower. Both lenders and borrowers can also enjoy better rates because such DeFi protocols have significantly lower operating costs than traditional banks. In addition, the metrics of these DeFi protocols are fully transparent—I can see in real time how much the protocol has in deposits and loans, and be assured that it is not over-leveraged.

What DeFi does for lending and borrowing, it can also do for many other financial services. Today, there are many dApps allowing users to trade assets (e.g. Uniswap and SushiSwap), speculate in futures or other derivatives (e.g. dYdX and Synthetix), and automate portfolio management (e.g. Yearn Finance and Set Protocol)—all without having to go through any intermediaries. In doing so, DeFi offers the promise of radically improving access to financial services, not only by reducing costs for users, but also allowing users to partake in more lucrative trading strategies without large upfront capital (or virtually none at all in the case of flash loans).

Of course, despite the benefits of DeFi, there are still significant risks in this space. The primary ones are smart contract risk (loopholes in the smart contract allowing for exploits), governance risk (malicious actors gaining control of a protocol and making changes that are not favourable to users), regulatory risk (governments deciding to outright ban or limit usage of DeFi protocols) and custodial risk (if you lose your private keys to your crypto wallet, your assets tied to it can no longer be accessed). That said, I do believe that these risks may be mitigated somewhat as the space matures, through better auditing of smart contracts, increased accessibility of DeFi insurance products, greater regulatory clarity from governments, and improved user design for crypto applications and wallets.

Aligning interests and facilitating cooperation

Another significant value proposition of blockchain-based applications is that by reducing intermediaries, ownership of the underlying protocols can be more equitably distributed among users. By aligning the interests of users more closely with that of the protocol as a whole, users become more likely to act or make decisions that can benefit the protocol (and thus themselves). The transparent nature of permissionless blockchains will also facilitate trust in the protocol itself, since users can be assured that any rules and incentives are fairly applied to all. All this can potentially open up new vistas of cooperation among people who are otherwise strangers.

The primary way that protocols distribute ownership among users is by rewarding users with governance tokens. Using the decentralised exchange SushiSwap as an example, users can earn SUSHI tokens by supplying liquidity to specific trading pairs. By then staking these SUSHI tokens within the application (which represents a commitment to hold on to those tokens rather than selling them), users are entitled to a share of the protocol’s profits from facilitating swaps of crypto assets. More importantly, each staked SUSHI token (xSUSHI) entitles the user to one vote on any governance proposal that will affect the protocol’s operations. Such governance proposals can include issues like deploying the protocol’s funds for various objectives (e.g. to boost liquidity for the protocol’s tokens on another dApp) or amending policy parameters (e.g. calibrating incentive amounts). Regardless, the central idea here is that by only allowing those who have staked their SUSHI tokens to vote, SushiSwap can ensure that the interests of its most committed users are aligned with the protocol—if SushiSwap grows and becomes successful, they benefit as well.

This mechanism of linking governance tokens to voting eligibility is the foundation of many DAOs. In fact, in its own words, SushiSwap is working towards becoming a DAO, with “working trustless governance”. There is no clear definition yet for what DAOs are, given their nascence. But we can broadly conceive of them as blockchain-enabled online collectives, in which the permissionless and transparent nature of blockchains are envisioned to empower groups of people to work together towards shared objectives without necessarily having to trust each other.

The range of DAOs that have sprung up is truly astounding, and examples include DAOs that aim to become a venture capital fund (e.g. the LAO), administer a crypto-collateralised stablecoin (e.g. MakerDAO), collect digital art (e.g. PleasrDAO), create an exclusive cultural community (e.g. Friends with Benefits), facilitate philanthropic grants (e.g. Big Green DAO) and even drive up the price of carbon offsets (e.g. Klima DAO).

Granted, not all DAOs will be havens of value creation or altruism. DAOs are not immune to the same pitfalls that afflict any human interactions, i.e. ego, greed, and what not. Within a DAO, the core leaders could still possibly collude to mislead the community, especially if there are aspects of governance and operations done off-chain. DAOs themselves could conceivably degenerate into siloed communities with socially divisive agendas too.

On a more practical note, the legal status of DAOs is also unclear, which may hinder their ability to do things in the “real” world. For instance, ConstitutionDAO recently managed to crowdfund over US$40 million worth of ether (ETH) to bid for a copy of the US Constitution that was put up for auction, but they had to set up a separate legal entity in order to do so (unfortunately, they did not win the auction).

Nevertheless, I am optimistic about the future of DAOs. The potential they offer in terms of facilitating trustless cooperation is immense and unprecedented. In particular, there is a compelling case for single purpose DAOs that can quickly bring together people with the right skills to achieve specific tasks. DAOs are also perfect sandboxes to test out socio-economic policies (e.g. the provision of universal basic income), since on-chain activities can be easily tracked and analysed in real time, and the scope of any trials can be rapidly scaled. I also believe these benefits will outweigh the risks over the long term. The web 3.0 community has a relatively strong ethos of openness and positivity, which I hope can be maintained as it grows. It is also likely there will be greater clarity on the legal status of DAOs in the future given their potential value—for a start, since July 2021, DAOs in Wyoming can have legal status as limited liability companies.

Evolving new ways of doing things

The final value proposition for blockchain-based applications, which is probably the most speculative out of the ones I have mentioned, is their potential to evolve entirely new ways of doing things. As Chris Dixon of a16z puts it in a blog post:

“New technologies enable activities that fall into one of two categories: 1) doing things you could already do but can now do better because they are faster, cheaper, easier, higher quality, etc. 2) doing brand new things that you simply couldn’t do before. Early in the development of new technologies, the first category tends to get more attention, but it’s the second that ends up having more impact on the world.”

Right now, many of the leading blockchain-based applications tend to be those that have clear analogs with existing applications. DeFi started by allowing you to engage financial services that you would already be familiar with from traditional finance, i.e. lending/borrowing, trading, and asset management. Many DAOs also have clear parallels to existing companies or non-government organisations, while most non-fungible tokens (NFTs) today are simply tokenised representations of art and media.

But I think the next frontier for blockchain-based applications is to create completely novel use cases that we could never have conceived before. Much like how developers can tap on application programming interfaces (APIs) to build ever-more powerful software products, blockchain-based applications are similarly, if not even more, composable. After all, the code of most dApps are open source, and participation in the leading protocols and blockchains are permissionless. Any developer in the blockchain space can easily iterate on these existing applications, recombine them or tear them apart, to unlock radically new possibilities.

A simple example of how composability can create new things is the concept of self-repaying loans, which initially blew my mind. This was pioneered by the Alchemix protocol, and the basic idea is that users can make loans that automatically pay themselves off, with no risk of liquidation. This screams almost fraudulent, until you realise such self-repaying loans simply build on existing DeFi protocols that allow you to earn yield on your collateral, which in turn can help pay off your loan over time. As I had earlier mentioned, some of these yields can actually be quite sustainable—for instance, the interest-generating loans offered by Compound and Aave will be always sufficiently collateralised. In any case, the outcome is simply more value unlocked for users, as illustrated in the example highlighted by Alchemix’s pseudonymous co-founder Scoopy Trooples in the following interview:

Stockhead: So what do people use the protocol for? Is it mostly just to cash out?

Trooples: We always like to hear stories of people using it for actual personal finance, I think that’s the most exciting stuff.

Probably the funniest story that we had with one of our users, is that a year ago, this guy’s dad’s boat sank because a hurricane came in, really roughed up the marina. And this user had done pretty well for himself in the crypto market. And what he did is he took his stablecoins, put them in Alchemix, took out an alUSD loan, and then used that to buy a boat for his dad.

Stockhead: Oh wow.

Trooples: So it’s like a self-repaying boat. And the advantage of doing this, as opposed to just taking out the stablecoins and buying it directly is, let’s say the boat costs $50,000, and you have $100,000 in stablecoins.

If you were to just liquidate that $50,000 and use that to buy the boat, you’re left with $50,000, and then you decide to put the rest of that $50,000 into earning some yield in DeFi, you would only have a principle base of $50,000 to do that. Whereas if you go to Alchemix you can kind of spend and save at the same time, because your full principle of $100,000 is earning yield.

So it’s effectively doubling your yield that you would get, as opposed to only having the $50,000 earning yield. But then with that extra $50,000 you have, you can go and spend it on whatever you want to, while your entire principle is making that money. And we’ve done some economic analysis of this, and it ends up being between a 15 to 20 per cent boost in capital efficiency.

Another arena where composability can possibly run wild is the world of NFTs. Right now, we largely think of them as discrete entities. But what if NFTs are like Lego blocks, and you can stack them in different ways to generate new products and services? For instance, I can imagine that such NFTs can help redefine how we think about CVs and resumes. Instead of linearly depicting one’s educational and career credentials in a flat document, we can use NFTs to present them in modular graphs that may more accurately represent how our experiences build on each other, or not—after all, career trajectories moving forward are no longer likely to run on a single path. Employers can also have greater assurance about the authenticity of one’s credentials, given that they cannot be tampered with once brought on-chain.

NFTs may also spawn entirely new ecosystems of characters, worlds, and even digital “life”. The project Loot simply started with various cards (minted as NFTs) containing just a list of items outlined in text, and many owners proceeded to build on their “loot” in myriad ways—be it to generate related artwork and stories, set up guilds based on individual items, and program smart contracts to unbundle their inventories. From this, it may not be hard to imagine that something akin to a Marvel Cinematic Universe could eventually evolve from just a collection of NFTs. I also actually think that NFTs can go much further, perhaps even be used to simulate life or any self-replicating systems digitally. A smart contract can be encoded within an NFT and programmed in a way that is analogous to a genome—these NFTs can then interact with each other to produce “offspring” (i.e. mint new NFTs) whose “genes” are a combination of their “parents”. Already, in the NFT-based game Axie Infinity, one can already breed Axies, which are digital creatures required to play the game. But I think the value of this can transcend simply having unique digital pets for entertainment. For instance, this can be used to run more complex ecological experiments that are untenable in the real world, in a permissionless and transparent manner.

Concluding thoughts

Ultimately, I believe the distributed, permissionless, immutable and transparent nature of blockchains open up many unique use cases that can add value to our lives:

• Blockchain-based applications can help reduce the need for intermediaries for some services, allowing us to cut down on middleman costs.

• Blockchains can also facilitate new ways of cooperation, enabling strangers to work together towards win-win outcomes.

• Finally, with permissionless composability, blockchains are likely to enable entirely new ways of doing things that lie beyond the current scope of our imagination.

I am hopeful about the technology and the communities that have sprung up around blockchains, and I hope that you are too after reading this piece. The breadth of the crypto space has broadened considerably in recent years, and I believe that there is now a place for anyone, whatever your background, to find a peg that can sustain your interests. In fact, I will add that a basic understanding of blockchain technology is likely to become essential in the future—over the next decade, financial literacy may require knowing how core DeFi protocols work, while keeping up with current affairs and sociopolitical developments may involve being aware of how DAOs function.

At the same time, I am also cautious about the risks, and pragmatic about the limitations. Blockchains will not necessarily displace everything that has come before. I am a crypto optimist, and not a crypto maximalist. I see blockchains as simply a new tool that can help augment the existing toolkit that we have steadily built up since the dawn of humanity. After all, the world is become ever more complex, and we will need as many good tools as possible in order to make it a better place for all of us. Wagmi.

Disclaimer: Nothing in this essay constitutes financial advice. Please do your own research and be mindful of your own investment objectives before investing into any cryptocurrencies, DeFi protocols or NFTs.

Cover image by Ezra Comeau from Pexels.